Professional Documents
Culture Documents
How Human Evolved
How Human Evolved
Fifth Edition
W. W. Norton & Company has been independent since its founding in 1923, when
William Warder Norton and Mary D. Herter Norton first published lectures delivered at
the People’s Institute, the adult education division of New York City’s Cooper Union.
The firm soon expanded its program beyond the Institute, publishing books by celebrated
academics from America and abroad. By mid-century, the two major pillars of Norton’s
publishing program—trade books and college texts—were firmly established. In the 1950s,
the Norton family transferred control of their company to its employees, and today—with
a staff of four hundred and a comparable number of trade, college, and professional titles
published each year—W. W. Norton & Company stands as the largest and oldest
publishing house owned wholly by its employees.
Copyright © 2009, 2006, 2003, 2000, 1997 by W. W. Norton & Company, Inc.
ISBN 978-0-393-93271-3
W. W. Norton & Company, Inc., 500 Fifth Avenue, New York, N.Y. 10110
wwnorton.com
W. W. Norton & Company, Ltd., Castle House, 75/76 Wells Street, London W1T 3QT
1 2 3 4 5 6 7 8 9 0
0a_Front Matter [4p]:How Humans Evolved [5e] 10/31/08 4:45 PM Page iii
HOW HUMANS
EVOLVED
CONTENTS
Preface xv
Chapter 2: Genetics 24
Mendelian Genetics 24
Cell Division and the Role of Chromosomes in Inheritance 26
Mitosis and Meiosis 27
Chromosomes and Mendel’s Experimental Results 28
Linkage and Recombination 32
viii CONTENTS
Population Genetics 53
Genes in Populations 54
How Random Mating and Sexual Reproduction Change
Genotypic Frequencies 55
CONTENTS ix
BOX 5.1 T EETH AND GUTS : YOU ARE WHAT YOU C AN C HEW 131
Activity Patterns 135
Ranging Behavior 136
Predation 138
Primate Sociality 141
x CONTENTS
CONTENTS xi
xii CONTENTS
CONTENTS xiii
xiv CONTENTS
Glossary A3
Credits A17
Index A23
0b_Preface [4p]:How Humans Evolved [5e] 10/31/08 4:47 PM Page xv
PREFACE
How Humans Evolved focuses on the processes that have shaped human evolution.
This approach reflects our training and research interests. As anthropologists, we are
interested in the evolutionary history of our own species, Homo sapiens. As evolu-
tionary biologists, we study how evolution works. In this book, we integrate these two
perspectives. We use current theoretical and empirical work in evolutionary theory,
population genetics, and behavioral ecology to interpret human evolutionary history.
We describe the changes that have occurred as the human lineage has evolved, and we
consider why these changes may have happened. We try to give life to the creatures
that left the bones and made the artifacts that paleontologists and archaeologists pain-
stakingly excavate by focusing on the processes that generate change, create adapta-
tions, and shape bodies and behavior. We also pay serious attention to the role of
evolution in shaping contemporary human behavior. There is considerable controversy
about evolutionary approaches to human behavior within the social sciences, but we
think it is essential to confront these issues openly and clearly. Positive responses to
the first four editions of How Humans Evolved tell us that many of our colleagues
endorse this approach.
One of the problems in writing a textbook about human evolution is that there
is considerable debate on many topics. Evolutionary biologists disagree about how
new species are formed and how they should be classified; primatologists argue about
the adaptive significance of infanticide and the effects of dominance rank on repro-
ductive performance; paleontologists disagree about the taxonomic relationships
among early hominin species and the emergence of modern humans; and those who
study modern humans disagree about the meaning and significance of race, the role
of culture in shaping human behavior and psychology, the adaptive significance of
many aspects of modern human behavior, and many other things. Sometimes multi-
ple interpretations of the same data can be defended; in other cases, the facts seem con-
tradictory. Textbook writers can handle this kind of uncertainty in two different ways.
They can weigh the evidence, present the ideas that best fit the available evidence, and
ignore the alternatives. Or they can present opposing ideas, evaluate the logic under-
lying each idea, and explain how existing data support each of the positions. We chose
the second alternative, at the risk of complicating the text and frustrating readers look-
ing for simple answers. We made this choice because we believe that this approach is
essential for understanding how science works. Students need to see how theories are
developed, how data are accumulated, and how theory and data interact to shape our
ideas about how the world works. We hope that students remember this long after they
have forgotten many of the facts that they will learn in this book.
xv
0b_Preface [4p]:How Humans Evolved [5e] 10/31/08 4:47 PM Page xvi
xvi Preface
sil finds that reveal new chapters in human ancestry, new data that alter our interpre-
tation of the behavioral strategies of primates, and experimental studies that reveal
cross-cultural regularities in mating preferences. New developments in human evolu-
tionary studies require regular updates of the textbook. Although we have made many
changes throughout the book, readers familiar with prior editions will find significant
changes in Part Two (“Primate Ecology and Behavior”), Part Three (“The History of
the Human Lineage”), and Part Four (“Evolution and Modern Humans”).
In Part One, we added an introduction to the molecular-genetics section of Chap-
ter 2, emphasizing the importance of molecular data in studying human evolution. We
also updated the molecular-genetics section, adding information about the roles of alter-
native splicing and microRNAs in gene expression. In Chapter 4 we added recently
published data on the extent of gene flow between species.
In Part Two, we combined and extensively revised Chapters 5 and 6 into one chap-
ter on primate diversity and ecology. We streamlined the discussions of primate tax-
onomy, dietary specializations, territoriality, and sociality, preserving the important
ideas and culling out distracting details. We also included new data on evidence for
predation in primates. In Chapter 6 we also added new material on the importance of
life history on variation in fitness, including an expanded discussion of sources of vari-
ation in female reproductive success that includes the effects of female age, longevity,
and quality of social bonds. There is an updated discussion of the relationship
between male rank and reproductive success as well. We also simplified the section on
sexual selection by limiting our discussion to processes likely to be important in pri-
mates and added new data on inbreeding avoidance. We changed the focus of Chap-
ter 7 to cooperation in order to allow a more focused analysis of primate social
behavior. We relocated the discussion of parent–offspring conflict from Chapter 6 to
Chapter 7 and integrated it into our discussion of kin selection and the conflicts of
interest among kin. We also updated our discussion on mutualism and altruism, giv-
ing mutualism a more prominent place in the chapter, and we included new informa-
tion on cooperative breeding primates. Chapter 8 includes new data regarding the link
among ecology, life history, and brain size in orangutans and comparisons of social/
nonsocial cognition in children and great apes. We streamlined our discussion of grade
changes in cognition over the course of primate evolution and added examples of pri-
mates’ knowledge of others’ minds.
We have updated Part Three to incorporate several major new findings. In Chap-
ter 9 we added a discussion of the recently discovered Miocene apes, including Piero-
pithecus catalaunicus. In Chapter 10, we adopted the term australopiths to allow
broad descriptions of hominin species that existed between 1 and 5 mya, and we
restored habilis to genus Homo. In the third edition, we had adopted Bernard Woods’s
classification that places the species habilis in the genus Australopithecus. We still think
this is sensible, but since there were lots of complaints, we put it back in Homo. Tax-
onomy is not really about the world but what to call things; so here we have bowed
to convention. We reorganized this chapter to discuss hominin species chronologically
(rather than in the order they were discovered). We also incorporated the new mate-
rial findings on A. anamensis and A. africanus, including a discussion of the Dikika
child, and added a revised discussion on australopith lifestyles based on new diet infor-
mation from tooth-enamel analyses. We added much new information to Chapter 12,
including discussions of diversity in H. ergaster, new information about H. ergaster’s
lifestyle—including a phylogeny of human tapeworms that shows evidence of early
meat eating in H. ergaster—and new postcranial material from Dmanisi. We have also
incorporated additional material on the controversy surrounding the classification of
0b_Preface [4p]:How Humans Evolved [5e] 10/31/08 4:47 PM Page xvii
Preface xvii
xviii Preface
For Instructors
CREATE DYNAMIC CLASSROOM PRESENTATIONS WITH THE
HOW HUMANS EVOLVED NORTON MEDIA LIBRARY CD-ROM
The ultimate tool for classroom presentation, this disc features nearly all the drawn
art and photos from the text in PowerPoint and JPEG format as well as a complete set
of PowerPoint lecture outlines for each chapter.
For Students
ORGANIZE, LEARN, AND CONNECT WITH STUDYSPACE
A WWNORTON.COM/STUDYSPACE
StudySpace is the free and open student website that accompanies the Fifth Edition of
How Humans Evolved. It offers assignment-driven study plans for each chapter that
help students Organize, Learn, and Connect with the course material and beyond.
• Organize—
Study plans help students organize their study and review
Chapter Reviews and Outlines
0b_Preface [4p]:How Humans Evolved [5e] 10/31/08 4:47 PM Page xix
Preface xix
• Learn—
Premium content ebook links
Chapter Quizzes
Flashcards, with audio pronunciation, allow students to test their knowledge of
important fossils and practice key terms.
• Connect—
Using the power of Google Earth™, AnthroTours take students on virtual field
trips to important anthropological sites around the globe. Each site includes
text and figures that help students interpret its most important elements.
EBOOK
ACKNOWLEDGMENTS
Over the last 10 years, many of our colleagues have provided new information, help-
ful comments, and critical perspectives that have enriched this book. We are grateful
for all those who have responded to our requests for photographs, clarifications, ref-
erences, and opinions. For the Fourth Edition, Laura MacLatchy provided help with
the Miocene apes in Chapter 10, Dan Fessler and David Schmitt gave us access to
material for Chapter 16, and Kermyt Anderson dug up original data for figures in
Chapter 17. Steven Reznik reviewed our discussion of the rapid evolution of placen-
tas in the minnows he studies and kindly provided an image. Leslie Aiello helped with
our discussion of hominin developmental rates. For help with the Third Edition, we
thank Carola Borries, Colin Chapman, Richard Klein, Cheryl Knott, Sally McBrearty,
Ryne Palombit, Steve Pinker, Karin Stronswold, and Bernard Wood. For help with the
Second Edition, we also thank Tom Plummer, Daniel Povinelli, Beverly Strassman, and
Patricia Wright. We remain grateful for the help we received for the First Edition from
Leslie Aiello, Monique Borgerhoff Mulder, Scott Carroll, Dorothy Cheney, Glenn Con-
roy, Martin Daly, Robin Dunbar, Lynn Fairbanks, Sandy Harcourt, Kristin Hawkes,
Richard Klein, Phyllis Lee, Nancy Levine, Jeff Long, Joseph Manson, Henry McHenry,
John Mitani, Jocelyn Peccei, Susan Perry, Steve Pinker, Tom Plummer, Tab Rasmussen,
Mark Ridley, Alan Rogers, Robert Seyfarth, Frank Sulloway, Don Symons, Alan
Walker, Tim White, and Margo Wilson.
A number of people reviewed all or parts of the first four editions. We thank the
following: Stephanie Anestis, Thad Bartlett, Rene Bobe, Barry Bogin, Doug Broadfield,
Margaret Clarke, Douglas Crews, Roberto Delgato, Arthur Durband, Charles
Edwards, Donald Gaff, Sharon Gursky, Mark Griffin, Kim Hill, Andrew Irvine, Bar-
bara King, Richard Klein, Darrell La Lone, Clark Larsen, Lynette Leidy, Laura
MacLatchy, Marilyn Norconk, Ann Palkovich, James Paterson, Michael Schillaci, Liza
Shapiro, Eric Smith, Craig Stanford, Horst Steklis, Joan Stevenson, Mark Stoneking,
0b_Preface [4p]:How Humans Evolved [5e] 10/31/08 4:47 PM Page xx
xx Preface
Rebecca Storey, Rebecca Stumpf, Timothy Weaver, and Patricia Wright. Several
anonymous reviewers also read previous editions and provided suggestions. Although
we are certain that we have not satisfied all those who read and commented on parts
of the book, we found all of the comments to be very helpful as we revised the text.
Richard Klein provided us with many exceptional drawings of fossils that appear
in Part Three—an act of generosity that we continue to appreciate. We also give spe-
cial thanks to Neville Agnew and the Getty Conservation Institute, which granted us
permission to use images of the Laetoli conservation project for the cover of the Sec-
ond Edition.
Many users of the book have commented on the quality of the illustrations. For
this we must thank the many friends and colleagues who allowed us to use their pho-
tographs: Bob Bailey, Carola Borries, Colin Chapman, Nick Blurton Jones, Sue Boin-
ski, Monique Borgerhoff Mulder, Richard Byrne, Scott Carroll, Marina Cords, Diane
Doran, Robert Gibson, Peter Grant, Kim Hill, Kevin Hunt, Lynne Isbell, Charles Jan-
son, Alex Kacelnik and the Behavioral Ecology Research Group, Nancy Levine, Car-
lão Limeira, Joe Manson, Frank Marlowe, Laura MacLatchy, Bill McGrew, John
Mitani, Claudio Nogueira, Ryne Palombit, Susan Perry, Craig Stanford, Karen Strier,
Alan Walker, Katherine West, and John Yellen. The National Museums of Kenya
kindly allowed us to reprint a number of photographs.
We also acknowledge the thousands of students and dozens of teaching assistants
at UCLA who have used various versions of this material over the years. Student eval-
uations of the original lecture notes, the first draft of the text, and the first four edi-
tions were helpful as we revised and rewrote various parts of the book. The teaching
assistants helped us identify many parts of the text that needed to be clarified, cor-
rected, or reconsidered.
We thank all the people at Norton who helped us produce this book, particularly
our outstanding editors Leo Wiegman and Pete Lesser. We are also grateful to our
excellent developmental editor, Carol Flechner. We would also like to thank all of those
who saw the book through the production process, including Eric Pier-Hocking, Thom
Foley, and Laura Musich. And special thanks to Elizabeth Erhart at Texas State Uni-
versity, San Marcos, for preparing the revised instructor’s manual; Kristina Killgrove
of the University of North Carolina, Chapel Hill, for preparing the revised StudySpace
review materials; and Sara Simon of the University of North Carolina, Chapel Hill, for
creating the audio pronunciation guide and AnthroTours.
0c_Prologue [3p]:How Humans Evolved [5e] 10/30/08 11:55 AM Page xxi
P ro l o g u e
I
n 1838, Charles Darwin discovered the principle of evolution by natural selec-
tion and revolutionized our understanding of the living world. Darwin was 28
years old, and it was just two years since he had returned from a five-year voy-
age around the world as a naturalist on the HMS Beagle (Figure 1). Darwin’s obser-
vations and experiences during the journey had convinced him that biological species
change through time and that new species arise by the transformation of existing ones,
xxi
0c_Prologue [3p]:How Humans Evolved [5e] 10/30/08 11:55 AM Page xxii
and he was avidly searching for an explanation of how these processes worked. In late
September of the same year, Darwin read Thomas Malthus’s Essay on the Principle of
Population, in which Malthus (Figure 2) argued that human populations invariably
grow until they are limited by starvation, poverty, and death. Darwin realized that
Malthus’s logic also applied to the natural world, and this intuition inspired the con-
ception of his theory of evolution by natural selection. In the intervening century and
a half, Darwin’s theory has been augmented by discoveries in genetics and amplified
by studies of the evolution of many types of organisms. It is now the foundation of our
understanding of life on Earth.
This book is about human evolution, and we will spend a lot of time explaining
how natural selection and other evolutionary processes have shaped the human
When this portrait species. Before we begin, it is important to consider why you should care about this
FIGURE 1 topic. Many of you will be working through this book as a requirement for an under-
of Charles Darwin
was painted, he was about 30 years old. graduate class in biological anthropology and will read the book in order to earn a
He had just returned from his voyage on good grade. As instructors of a class like this ourselves, we approve of this motive.
the HMS Beagle and was still busy However, there is a much better reason to care about the processes that have shaped
organizing his notes, drawings, and vast human evolution: understanding how humans evolved is the key to understanding why
collections of plants and animals. people look and behave the way they do.
The profound implications of evolution for our understanding of humankind were
apparent to Darwin from the beginning. We know this today because he kept note-
books in which he recorded his private thoughts about various topics. The quotation
that begins this prologue is from the M Notebook, begun in July 1838, in which Dar-
win jotted down his ideas about humans, psychology, and the philosophy of science.
In the nineteenth century, metaphysics involved the study of the human mind. Thus,
Darwin was saying that because he believed humans evolved from a creature some-
thing like a baboon, it followed that an understanding of the mind of a baboon would
contribute more to an understanding of the human mind than would all of the works
of the great English philosopher John Locke.
Darwin’s reasoning was simple. Every species on this planet has arisen through the
same evolutionary processes. These processes determine why organisms are the way
they are by shaping their morphology, physiology, and behavior. The traits that char-
acterize the human species are the result of the same evolutionary processes that cre-
ated all other species. If we understand these processes and the conditions under which
the human species evolved, then we will have the basis for a scientific understanding
of human nature. Trying to comprehend the human mind without an understanding
of human evolution is, as Darwin wrote in another notebook that October, “like puz-
zling at astronomy without mechanics.” By this, Darwin meant that his theory of evo-
lution could play the same role in biology and psychology that Isaac Newton’s laws
of motion had played in astronomy. For thousands of years, stargazers, priests,
philosophers, and mathematicians had struggled to understand the motions of the
planets without success. Then, in the late 1600s, Newton discovered the laws of
mechanics and showed how all of the intricacies in the dance of the planets could be
explained by the action of a few simple processes (Figure 3).
In the same way, understanding the processes of evolution enables us to account
Thomas Malthus was
FIGURE 2 for the stunning sophistication of organic design and the diversity of life, and to under-
the author of An
Essay on the Principle of Population, a stand why people are the way they are. As a consequence, understanding how natu-
book Charles Darwin read in 1838 that ral selection and other evolutionary processes shaped the human species is relevant to
profoundly influenced the development all of the academic disciplines that are concerned with human beings. This vast intel-
of his theory of evolution by natural lectual domain includes medicine, psychology, the social sciences, and even the
selection. humanities. Beyond academia, understanding our own evolutionary history can help
0c_Prologue [3p]:How Humans Evolved [5e] 10/30/08 11:55 AM Page xxiii
us answer many questions that confront us in everyday life. Some of these questions
are relatively trivial: Why do we sweat when hot or nervous? Why do we crave salt,
sugar, and fat, even though large amounts of these substances cause disease (Figure 4)?
Why are we better marathon runners than mountain climbers? Other questions are
more profound: Why do only women nurse their babies? Why do we grow old and
eventually die? Why do people look so different around the world? As you will see,
evolutionary theory provides answers or insights about all of these questions. Aging,
which eventually leads to death, is an evolved characteristic of humans and most other
creatures. Understanding how natural selection shapes the life histories of organisms
tells us why we are mortal, why our life span is about 70 years, and why other species
live shorter lives. In an age of horrific ethnic conflicts and growing respect for multi-
cultural diversity, we are constantly reminded of the variation within the human Sir Isaac Newton
species. Evolutionary analyses tell us that genetic differences between human groups FIGURE 3
discovered the laws
are relatively minor, and that our notions of race and ethnicity are culturally con- of celestial mechanics, a body of theory
structed categories, not biological realities. that resolved age-old mysteries about
All of these questions deal with the evolution of the human body. However, under- the movements of the planets.
standing evolution is also an important part of our understanding of human behav-
ior and the human mind. The claim that understanding evolution will help us
understand contemporary human behavior is much more controversial than the
claim that it will help us understand how human bodies work. But it should not be.
The human brain is an evolved organ of great complexity, just like the endocrine sys-
tem, the nervous system, and all of the other components of the human body that reg-
ulate our behavior. Understanding evolution helps us understand our mind and
behavior because evolutionary processes forged the brain that controls human behav-
ior, just as they forged the brain of the chimpanzee and the salamander.
One of the great debates in Western thought centers on the essence of human
nature. One view is that people are basically honest, generous, and cooperative crea-
tures who are corrupted by an immoral economic and social order. The opposing view
is that we are fundamentally amoral, egocentric beings whose antisocial impulses are
held in check by social pressures. This question turns up everywhere. Some people
believe that children are little barbarians who are civilized
only through sustained parental effort; others think that chil-
dren are gentle beings who are socialized into competitive-
ness and violence by exposure to negative influences like toy
guns and violent TV programs (Figure 5). The same di-
chotomy underpins much political and economic thought.
Economists believe that people are rational and selfish, but
other social scientists, particularly anthropologists and soci-
ologists, question and sometimes reject this assumption. We
can raise an endless list of interesting questions about human
nature: Does the fact that, in most societies, women rear chil-
dren and men make war mean that men and women differ
in their innate predispositions? Why do men typically find
younger women attractive? Why do some people neglect and
abuse their children, while others adopt and lovingly raise
A strong appetite for sugar, fat, and salt may have
children who are not their own? FIGURE 4
been adaptive for our ancestors, who had little
Understanding human evolution does not reveal the access to sweet, fatty, and salty foods. We have inherited these
answers to all of these questions or even provide a complete appetites and have easy access to these foods. As a consequence,
answer to any one of them. As we will see, however, it can many of us suffer from obesity, high blood pressure, diabetes, and
provide useful insights about all of them. An evolutionary heart disease.
0c_Prologue [3p]:How Humans Evolved [5e] 10/30/08 11:55 AM Page xxiv
approach does not imply that behavior is “genetically determined” or that learning
and culture are unimportant. In fact, we will argue that learning and culture play cru-
cial roles in human behavior. Behavioral differences among peoples living in different
times and places result mainly from flexible adjustments to different social and envi-
ronmental conditions. Understanding evolution is useful precisely because it helps us
understand why humans respond in different ways to different conditions.
P a rt O n e
HOW EVOLUTION
WORKS
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 2
Chapter 1
Explaining Adaptation before Darwin
Darwin’s Theory of Adaptation
Darwin’s Postulates
An Example of Adaptation
by Natural Selection
Individual Selection
The Evolution of Complex Adaptations
Why Small Variations Are Important
Why Intermediate Steps Are
Favored by Selection
Rates of Evolutionary Change
Darwin’s Difficulties Explaining Variation
ADAPTATION BY
NATURAL SELECTION
E
ven the casual observer can see that organisms are well suited to their cir-
cumstances. For example, fish are clearly designed for life under water, and
certain flowers are designed to be pollinated by particular species of insects.
More careful study reveals that organisms are more than just suited to their envi-
ronments: they are complex machines, made up of many exquisitely constructed com-
ponents, or adaptations, that interact to help the organism survive and reproduce.
2
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 3
The human eye provides a good example of an adap- Transparent liquid Transparent jelly
tation. Eyes are amazingly useful: they allow us to move
confidently through the environment, to locate critical
resources like food and mates, and to avoid dangers like Lens Retina
predators and cliffs. Eyes are extremely complex struc-
tures made up of many interdependent parts (Figure
Cornea
1.1). Light enters the eye through a transparent opening,
then passes through a diaphragm called the “iris,” which
regulates the amount of light entering the eye and allows Iris
the eye to function in a wide range of lighting conditions.
The light then passes through a lens that projects a
focused image on the retina on the back surface of the Muscles
eye. Several different kinds of light-sensitive cells then Optic nerve
convert the image into nerve impulses that encode infor-
A cross section of the human eye.
mation about spatial patterns of color and intensity. FIGURE 1.1
These cells are more sensitive to light than the best pho-
tographic film is. The detailed construction of each of these parts of the eye makes
sense in terms of the eye’s function: seeing. If we probed into any of these parts, we
would see that they, too, are made of complicated, interacting components whose
structure is understandable in terms of their function.
Differences between human eyes and the eyes of other animals make sense in terms
of the types of problems each creature faces. Consider, for example, the eyes of fish
and humans (Figure 1.2). The lens in the eyes of humans and other terrestrial mam-
mals is much like a camera lens; it is shaped like a squashed football and has the same
index of refraction (a measure of light-bending capacity) throughout. In contrast, the
lens in fish eyes is a sphere located at the center of the curvature of the retina, and the
index of refraction of the lens increases smoothly from the surface of the lens to the
center. It turns out that this kind of lens, called a “spherical gradient lens,” provides
a sharp image over a full 180° visual field, a very short focal length, and high light-
gathering power—all desirable properties. Terrestrial creatures like us cannot use this
design because light is bent when it passes from the air through the cornea (the trans-
parent cover of the pupil), and this fact constrains the design of the remaining lens ele-
ments. In contrast, light is not bent when it passes from water through the cornea of
aquatic animals, and the design of their eyes takes advantage of this fact.
Before Darwin there was no scientific explanation for the fact that organisms are
well adapted to their circumstances.
Charles Darwin was born into a well-to-do, intellectual, and politically liberal family
in England. Like many prosperous men of his time, Darwin’s father wanted his son to
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 5
become a doctor. But after failing at the prestigious medical school at the University
of Edinburgh, Charles went on to Cambridge University, resigned to becoming a coun-
try parson. He was, for the most part, an undistinguished student—much more inter-
ested in tramping through the fields around Cambridge in search of beetles than in
studying Greek and mathematics. After graduation, one of Darwin’s botany profes-
sors, John Stevens Henslow, provided him with a chance to pursue his passion for nat-
ural history as a naturalist on the HMS Beagle.
The Beagle was a Royal Navy vessel whose charter was to spend two to three
years mapping the coast of South America and then to return to London, perhaps by
circling the globe (Figure 1.4). Darwin’s father forbade him to go, preferring that he
get serious about his career in the church, but Darwin’s uncle (and future father-in-law)
Josiah Wedgwood II intervened. The voyage turned out to be the turning point in Dar-
win’s life. His work during the voyage established his reputation as a skilled natural-
ist. His observations of living and fossil animals ultimately convinced him that plants
and animals sometimes change slowly through time, and that such evolutionary
change is the key to understanding how new species come into existence. This view
was rejected by most scientists of the time and was considered heretical by the gen-
eral public.
Darwin’s Postulates
Darwin’s theory of adaptation follows from three postulates: (1) the struggle
for existence, (2) variation in fitness, and (3) the inheritance of variation.
In 1838, shortly after the Beagle returned to London, Darwin formulated a simple
mechanistic explanation for how species change through time. His theory follows from
three postulates:
1. The ability of a population to expand is infinite, but the ability of any environ-
ment to support populations is always finite.
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 6
2. Organisms within populations vary, and this variation affects the ability of indi-
viduals to survive and reproduce.
3. This variation is transmitted from parents to offspring.
Darwin’s first postulate means that populations grow until they are checked by the
dwindling supply of resources in the environment. Darwin referred to the resulting
competition for resources as “the struggle for existence.” For example, animals
require food to grow and reproduce. When food is plentiful, animal populations grow
until their numbers exceed the local food supply. Because resources are always finite,
it follows that not all individuals in a population will be able to survive and reproduce.
According to the second postulate, some individuals will possess traits that enable
them to survive and reproduce more successfully (producing more offspring) than oth-
ers in the same environment. The third postulate holds that if the advantageous traits
are inherited by offspring, then these traits will become more common in succeeding
generations. Thus, traits that confer advantages in survival and reproduction are
retained in the population, and traits that are disadvantageous disappear. When Dar-
win coined the term natural selection for this process, he was making a deliberate anal-
ogy to the artificial selection practiced by animal and plant breeders of his day. A much
more apt term would be “evolution by variation and selective retention.”
In his autobiography, first published in 1887, Darwin claimed that the curious pattern
of adaptations he observed among the several species of finches that live on the Galá-
pagos Islands off the coast of Ecuador—now referred to as “Darwin’s finches”—was
crucial in the development of his ideas about evolution (Figure 1.5). Recently discov-
ered documents suggest that Darwin was actually quite confused about the Galápa-
(a)
(b)
(a) The islands of the Galápagos, which are located off the coast of Ecuador,
FIGURE 1.5
house a variety of unique species of plants and animals. (b) Cactus finches from
Charles Darwin’s The Zoology of the Voyage of H.M.S. Beagle (1840).
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 7
gos finches during his visit, and they played little role in his
discovery of natural selection. Nonetheless, Darwin’s finches
hold a special place in the minds of most biologists.
Peter and Rosemary Grant, biologists at Princeton Uni-
versity, conducted a landmark study of the ecology and evo-
lution of one particular species of Darwin’s finches on one of
the Galápagos Islands. The study is remarkable because the
Grants were able to directly document how Darwin’s three
postulates lead to evolutionary change. The island, Daphne
Major, is home to the medium ground finch (Geospiza for-
tis), a small bird that subsists mainly by eating seeds (Figure
1.6). The Grants and their colleagues caught, measured,
weighed, and banded nearly every finch on the island each
year of their study—some 1,500 birds in all. They also kept The medium ground finch, Geospiza fortis, uses
FIGURE 1.6
track of critical features of the birds’ environment, such as its beak to crack open seeds. (Photograph
the distribution of seeds of various sizes, and they observed courtesy of Peter Grant.)
the birds’ behavior.
A few years into the Grants’ study, a severe drought struck Daphne Major (Fig-
ure 1.7). During the drought, plants produced many fewer seeds, and the finches soon
depleted the stock of small, soft, easily processed seeds, leaving only large, hard,
difficult-to-process seeds (Figure 1.8). The bands on the birds’ legs enabled the Grants
to track the fate of individual birds during the drought, and the regular measurements
that they had made of the birds allowed them to compare the traits of birds that
survived the drought with the traits of those that perished. The Grants also kept de-
tailed records of the environmental conditions, which allowed them to determine how
the drought affected the finch’s habitat. It was this vast body of data that enabled the
Grants to document the action of natural selection among the finches of Daphne
Major.
The Grants’ data show how the processes identified in Darwin’s postulates lead
to adaptation.
The events on Daphne Major embodied all three of Darwin’s postulates. First, the
supply of food on the island was not sufficient to feed the entire population, and many
finches did not survive the drought. From the beginning of the drought in 1976 until
Daphne Major in
FIGURE 1.7
March 1976 (a),
after a year of good rains; and in March
1977 (b), after a year of very little
rain.
(a) (b)
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 8
Drought the rains came nearly two years later, the population of
medium ground finches on Daphne Major declined from
1,200 birds to only 180.
Second, beak depth (the top-to-bottom dimension of
Index of seed size and hardness
the beak) varied among the birds on the island, and this
variation affected the birds’ survival. Before the drought
began, the Grants and their colleagues had observed that
birds with deeper beaks were able to process large, hard
seeds more easily than birds with shallower beaks were.
Deep-beaked birds usually concentrated on large seeds,
while shallow-beaked birds normally focused their efforts
on small seeds. The open bars in the histogram in Figure
1.9a show what the distribution of beak sizes in the popu-
lation was like before the drought. The height of each open
bar represents the number of birds with beaks in a given
range of depths—for example, 8.8 to 9.0 mm, or 9.0 to 9.2
1976 1977 1978 mm. During the drought, the relative abundance of small
Year seeds decreased, forcing shallow-beaked birds to shift to
During the two-year drought, the size and hard- larger and harder seeds. Shallow-beaked birds were then at
FIGURE 1.8 a distinct disadvantage because it was harder for them to
ness of seeds available on Daphne Major
increased because birds consumed all of the desirable small, soft seeds, crack these seeds. The distribution of individuals within the
leaving mainly larger and harder seeds. Each point on this plot repre- population changed during the drought because finches
sents an index of seed size and hardness at a given time. with deeper beaks were more likely to survive than finches
with shallow beaks (Figure 1.9b). The shaded portion of
the histogram in Figure 1.9a shows what the distribution of beak depths would have
been like among the survivors. Because many birds died, there were fewer remaining
in each category. However, mortality was quite specific. The proportion of shallow-
Frequency of each beak depth
Probability of survival
Before selection
After
selection
beaked birds that died greatly exceeded the proportion of deep-beaked birds that died.
As a result, the shaded portion of the histogram shows a shift to the right, which means
that the average beak depth in the population increased. Thus the average beak depth
among the survivors of the drought was greater than the average beak depth in the
same population before the drought.
Third, parents and offspring had similar beak depths. The Grants discovered this
by capturing and banding nestlings and recording the identity of the nestlings’ parents.
When the nestlings became adults, the Grants recaptured and measured them. The
Grants found that, on average, parents with deep beaks produced offspring with deep
beaks (Figure 1.10). Because parents were drawn from the pool of individuals who sur-
vived the drought, their beaks were, on average, deeper than those of the original res-
idents of the island; and because offspring resemble their parents, the average beak
depth of the survivors’ offspring was greater than the aver-
age beak depth before the drought. This means that,
Drought
through natural selection, the average morphology (an
organism’s size, shape, and composition) of the bird popu-
lation changed so that birds became better adapted to their
environment. This process, operating over approximately
two years, led to a 4% increase in the mean beak depth in
Mean beak depth
Probability of survival
When birds with the most common beak depth are most likely to survive and
FIGURE 1.12
reproduce, natural selection keeps the mean beak depth constant. (a) Birds
with deep or shallow beaks are less likely to survive than birds with average beaks. Birds with shallow
beaks cannot process large, hard seeds; and birds with deep beaks are less likely to survive to adulthood.
(b) The open bars represent the distribution of beak depths before selection, and the shaded bars rep-
resent the distribution after selection. As in Figure 1.9, notice that there are fewer birds in the popula-
tion after selection. Because birds with average beaks are most likely to survive, however, the peak of
the distribution of beak depths is not shifted and mean beak depth remains unchanged.
(Figure 1.12). Evolutionary theory predicts that, over time, selection will increase the
average beak depth in the population until the costs of larger-than-average beak size
exceed the benefits. At this point, finches with the average beak size in the population
will be the most likely to survive and reproduce, and finches with deeper or shallower
beaks than the new average will be at a disadvantage. When this is true, beak size does
not change, and we say that an equilibrium exists in the population with regard to
beak size. The process that produces this equilibrium state is called stabilizing selec-
tion. Notice that even though the average characteristics of the beak in the population
will not change in this situation, selection is still going on. Selection is required to
change a population, and selection is also required to keep a population the same.
It might seem that beak depth would also remain unchanged if this trait had no
effect on survival (or put another way, if there were no selection favoring one type of
beak over another). Then all types of birds would be equally likely to survive from one
generation to the next, and beak depth would remain constant. This logic would be
valid if selection were the only process affecting beak size. However, real populations
are also affected by other processes that cause traits, or characters, to change in unpre-
dictable ways. We will discuss these processes further in Chapter 3. The point to
remember here is that populations do not remain static over the long run unless selec-
tion is operating.
Beak size
Island.
Individual Selection
Adaptation results from the competition among individuals, not between entire
populations or species.
survive if all females produced fewer offspring. This does not happen, however, because
natural selection among individuals favors females who produce many offspring.
To see why selection on individuals will lead to this result, let’s consider a simple
hypothetical case. Suppose the females of a particular species of monkey are maxi-
mizing individual reproductive success when they produce 10 offspring. Females that
produce more than or less than 10 offspring will tend to leave fewer descendants in
the next generation. Further suppose that the likelihood of the species becoming
extinct would be lowest if females produced only two offspring apiece. Now suppose
that there are two kinds of females. Most of the population is composed of low-
fecundity females that produce just two offspring each, but a few high-fecundity
females produce 10 offspring each. (Fecundity is the term demographers use for the
ability to produce offspring.) High-fecundity females have high-fecundity daughters,
and low-fecundity females have low-fecundity daughters. The proportion of high-
fecundity females will increase in the next generation because such females produce
more offspring than do low-fecundity females. Over time, the proportion of high-
fecundity females in the population will increase rapidly. As fecundity increases, the
population will grow rapidly and may deplete available resources. The depletion of
resources, in turn, will increase the chance that the species becomes extinct. However,
this fact is irrelevant to the evolution of fecundity before the extinction, because nat-
ural selection results from competition among individuals, not competition among
species.
The idea that natural selection operates at the level of the individual is a key ele-
ment in understanding adaptation. In discussing the evolution of social behavior in
Chapter 7, we will encounter several additional examples of situations in which selec-
tion increases individual success but decreases the competitive ability of the population.
It was known in Darwin’s day that most variation is continuous. An example of con-
tinuous variation is the distribution of heights in people. Humans grade smoothly from
one extreme to the other (short to tall), with all the intermediate types (in this case,
heights) represented. However, Darwin’s contemporaries also knew about discontin-
uous variation, in which a number of distinct types exist with no intermediates. In
humans, height is also subject to discontinuous variation. For example, there is a
genetic condition, called achondroplasia, that causes affected individuals to be much
shorter than other people, have proportionately shorter arms and legs, and bear a vari-
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 13
ety of other distinctive features. Discontinuous variants are usually quite rare in nature.
Nonetheless, many of Darwin’s contemporaries who were convinced of the reality of
evolution believed that new species arise as discontinuous variants.
1 1 1 1
3 3 3z z z3 ø 10240
27 27 27 27
28 times
This is a very small number. To get a feeling for how small a chance there is of the
monkeys typing the sentence correctly, suppose a very fast computer (faster than any
currently in existence) could generate 100 billion (1011) characters per second and run
for the lifetime of the Earth—about 4 billion years, or 1017 seconds. Then the chance
of randomly typing the line “Methinks it is like a weasel” even once during the whole
of Earth’s history would be about 1 in 1 trillion! Typing the whole play is obviously
astronomically less likely; and although Hamlet is a very complicated thing, it is much
less complicated than a human eye. There’s no chance that a structure like the human
eye would arise by chance in a single trial. If it did, it would be, as the astrophysicist
Sir Frederick Hoyle said, “like a hurricane blowing through a junk yard and chanc-
ing to assemble a Boeing 747.”
Darwin argued that continuous variation is essential for the evolution of complex
adaptations. Once again, Richard Dawkins provides an example that makes clear
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 14
Darwin’s reasoning. Again imagine a room full of monkeys and typewriters, but now
the rules of the game are different. The monkeys type the first 28 characters at ran-
dom, and then during the next round they attempt to copy the same initial string of
letters and spaces. Most of the sentences are just copies of the previous string, but
because monkeys sometimes make mistakes, some strings have small variations, usu-
ally in only a single letter. During each trial, the monkey trainer selects the string that
most resembles Shakespeare’s phrase “Methinks it is like a weasel” as the string to be
copied by all the monkeys in the next trial. This process is repeated until the monkeys
come up with the correct string. Calculating the exact number of trials required to gen-
erate the correct sequence of characters is quite difficult, but it is easy to simulate the
process on a computer. Here’s what happened when Dawkins performed the simula-
tion. The initial random string was
WDLMNLT DTJBKWIRZREZLMQCO P
WDLMNLT DTJBSWIRZREZLMQCO P
After 10 trials:
After 20 trials:
After 30 trials:
After 40 trials:
The exact phrase was reached after 43 trials. Dawkins reports that it took his 1985-
vintage Macintosh only 11 seconds to complete this task.
Selection can give rise to great complexity starting with small random variations
because it is a cumulative process. As the typing monkeys show us, it is spectacularly
unlikely that a single random combination of keystrokes will produce the correct sen-
tence. However, there is a much greater chance that some of the many small random
changes will be advantageous. The combination of reproduction and selection allows
the typing monkeys to accumulate these small changes until the desired sentence is
reached.
There is a potent objection to the example of the typing monkeys. Natural selection,
acting over time, can lead to complex adaptations, but it can do so only if each small
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 15
change along the way is itself adaptive. Although it is easy to assume that this is true
in a hypothetical example of character strings, many people have argued that it is
unlikely for every one of the changes necessary to assemble a complex organ like the
eye to be adaptive. An eye is useful, it is claimed, only after all parts of the complex-
ity have been assembled; until then, it is worse than no eye at all. After all, what good
is 5% of an eye?
Darwin’s answer, based on the many adaptations for seeing or sensing light that
exist in the natural world, was that 5% of an eye is often better than no eye at all. It
is quite possible to imagine that a very large number of small changes—each favored
by selection—led cumulatively to the wonderful complexity of the eye. Living mol-
lusks, which display a broad range of light-sensitive organs, provide examples of many
of the likely stages in this process (Figure 1.15):
4. If the depression got deep enough (Figure 1.15c), it could form images on the
light-sensitive tissue, much the way pinhole cameras form images on photo-
graphic film. In settings in which detailed images are useful, selection could then
favor the elaboration of the neural machinery necessary to interpret the image.
5. The next step is the formation of a transparent cover (Figure 1.15d). This might
be favored because it protects the interior of the eye from parasites and mechan-
ical damage.
6. A lens could evolve through gradual modification of either the transparent cover
or the internal structures within the eye (Figure 1.15e, f).
Notice that evolution produces adaptations like a tinkerer, not an engineer. New
organisms are created by small modifications of existing organisms, not by starting
with a clean slate. Clearly many beneficial adaptations will not arise because they are
blocked at some step along the way when a particular variation is not favored by selec-
tion. Darwin’s theory explains how complex adaptations can arise through natural
processes, but it does not predict that every possible adaptation, or even most, will
occur. This is not the best of all possible worlds; it is just one of many possible worlds.
The fact that natural selection constructs complex adaptations like a tinkerer
might lead you to think that the assembly of complex adaptations is a chancy busi-
ness. If even a single step were not favored by selection, the adaptation could not arise.
Such reasoning suggests that complex adaptations are mere coincidence. Although
chance does play a very important role in evolution, the power of cumulative natural
selection should not be underestimated. The best evidence that selection is a power-
ful process for generating complex adaptations comes from a phenomenon called con-
vergence, the evolution of similar adaptations in unrelated groups of animals.
The similarity between the marsupial faunas of Australia and South America and
the placental faunas of the rest of the world provides a good example of convergence.
In most of the world, the mammalian fauna is dominated by placental mammals,
which nourish their young in the uterus during long pregnancies. Both Australia and
South America, however, became separated from an ancestral supercontinent, known
as Pangaea, long before placental mammals evolved. In Australia and South America,
marsupials (nonplacental mammals, like kangaroos, that rear their young in external
pouches) came to dominate the mammalian fauna, filling all available mammalian
niches. Some of these marsupial mammals were quite similar to the placental mammals
on the other continents. For example, there was a marsupial wolf in Australia that
looked very much like placental wolves of Eurasia, even sharing subtle features of their
feet and teeth (Figure 1.16). These marsupial wolves became extinct in the 1930s. Sim-
ilarly, in South America a marsupial saber-toothed cat independently evolved many of
the same adaptations as the placental saber-toothed cat that stalked North America
10,000 years ago. These similarities are more impressive when you consider that the
last common ancestor of marsupial and placental mammals was a small, nocturnal,
insectivorous creature, something like a shrew, that lived about 120 million years ago
(mya). Thus, selection transformed a shrew step by small step, each step favored by
selection, into a saber-toothed cat—and it did it twice. This cannot be coincidence.
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 17
The marsupial wolf that lived in Tasmania until early in the twentieth century
FIGURE 1.16
(drawn from a photograph of one of the last living animals). Similarities with
placental wolves of North America and Eurasia illustrate the power of natural selection to create com-
plex adaptations. Their last common ancestor was probably a small insectivorous creature like the shrew.
Geospiza fortis
Geospiza fuliginosa
Geospiza magnirostris
Geospiza difficilis
Geospiza conirostris
Geospiza scandens
Camarhynchus pauper
We can trace
FIGURE 1.18
the relationship Camarhynchus parvulus
among various species of Darwin’s
finches by analyzing their protein poly-
morphisms. Species that are closely Cactospiza
linked in the phylogenetic tree are more
similar to one another genetically than
Platyspiza
to other species because they share a
more recent common ancestor. The tree
does not include 3 of the 14 species of Certhidea
Darwin’s finches.
enough time for natural selection to produce the adaptations that we observe. The real
puzzle is why the change observed in the fossil record was so slow.
The Grants’ observation of the evolution of beak morphology in Darwin’s finches
provides one example of rapid evolutionary change. The medium ground finch of
Daphne Major is one of 14 species of finches that live in the Galápagos. Evidence sug-
gests that all 14 are descended from a single South American species that migrated to
the newly emerged islands about half a million years ago (Figure 1.18). This doesn’t
seem like a very long time. Is it possible that natural selection created 14 species in only
half a million years?
To start to answer the question, let’s calculate how long it would take for the
medium ground finch (Geospiza fortis) to come to resemble its closest relative, the
large ground finch (Geospiza magnirostris), in beak size and weight (Figure 1.19). The
large ground finch is 75% heavier than the medium ground finch, and its beak is about
20% deeper. Remember that beak size increased about 4% in two years during the
1977 drought. The Grants’ data indicate that body size also increased by a similar
amount. At this rate, Peter Grant calculated that it would take between 30 and 46
years for selection to increase the beak size and body weight of the medium ground
finch to match those of the large ground finch. But these changes occurred in response
to an extraordinary environmental crisis. The data suggest that selection doesn’t gen-
erally push consistently in one direction. Instead, in the Galápagos, evolutionary
change seems to go in fits and starts, moving traits one way and then another. So let’s
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 19
suppose that a net change in beak size like the one that occurred during 1977 occurs
only once every century. Then it would take about 2,000 years to transform the
medium ground finch into the large ground finch—still a very rapid process.
Similar rates of evolutionary change have been observed elsewhere when species
invade new habitats. For example, about 100,000 years ago a population of elk (called
“red deer” in Great Britain) colonized and then became isolated on the island of
Jersey, off the French coast, presumably by rising sea levels. By the time the island
was reconnected with the mainland approximately 6,000 years later, the red deer
had shrunk to about the size of a large dog. University of Michigan paleontologist
Philip Gingerich compiled data on the rate of evolutionary change in 104 cases in
which species invaded new habitats. These rates ranged from a low of zero (that
is, no change) to a high of about 22% per year, with an average of about 0.1% per The large
year. FIGURE 1.19
ground finch
The changes the Grants observed in the medium ground finch are relatively sim- (Geospiza magnirostris) has a beak that
ple: birds and their beaks just got bigger. More complex changes usually take longer is nearly 20% deeper than the beak of
to evolve, but several kinds of evidence suggest that selection can produce big changes its close relative, the medium ground
in remarkably short periods of time. finch (Geospiza fortis). At the rate of
One line of evidence comes from artificial selection. Humans have performed evolution observed during the drought
selection on domesticated plants and animals for thousands of years, and for most of of 1977, Peter Grant calculated that
this period this selection was not deliberate. There are many familiar examples. All selection could transform the medium
ground finch into the large ground finch
domesticated dogs, for instance, are believed to be descendants of wolves. Scientists
in less than 46 years. (Photograph
are not sure when dogs were domesticated, but 15,000 years ago is a good guess,
courtesy of Peter and Rosemary
which means that in a few thousand generations, selection changed wolves into
Grant.)
Pekingese, beagles, greyhounds, and Saint Bernards. In reality, most of these breeds
were created fairly recently, and most were the products of directed breeding. Darwin’s
favorite example of artificial selection was the domestication of pigeons. In the nine-
teenth century, pigeon breeding was a popular hobby, especially among working peo-
ple who competed to produce showy birds (Figure 1.20). They created a menagerie of
wildly different forms, all descended from the rather plain-looking rock pigeon. Dar-
win pointed out that these breeds are so different that if they were discovered in nature,
biologists would surely classify them as members of different species. Yet they were
produced by artificial selection within a few hundred years.
Rapid evolution of a complex feature has also been documented in a recent
study of a group of very closely related species of fish from the genus Poeciliopsis
(Figure 1.21). These small minnows can be found in tropical lowland streams, high-
altitude lakes, and desert springs and streams in Mexico and Central America. All
the species in this genus bear live young, but the sequence of events between fertil-
ization and birth varies. In most species, females endow the eggs with nutrients
before fertilization. As the young develop, they consume this endowment. As a con-
sequence, the emerging offspring are smaller than the egg they came from. In a few
species, however, females continue to provide nutrients to their unborn offspring
throughout development using highly differentiated tissues that are analogous to
mammalian placentas. When they are born, these offspring can be more than 100
times the mass of the egg at fertilization. Biologist David Reznick of the University
of California Riverside and his colleagues have shown that these placental tissues
evolved independently in three different groups of species within the genus Poecil-
iopsis. Genetic data indicate that one of these species groups diverged from ances-
tors lacking placental tissue only 0.75 mya, and the other two diverged less than 2.4
mya. These time estimates actually represent the time since these species shared a
common ancestor, and they set an upper bound on the amount of time required for
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 20
(b)
(a)
(c) (d)
the placenta to evolve. The generation time for these fish ranges from six months to
a year, so this complex adaptation evolved in fewer than a million generations.
A third line of evidence comes from theoretical studies of the evolution of com-
plex characters. Dan-Eric Nilsson and Susanne Pelger of
Lund University, in Sweden, have built a mathematical
model of the evolution of the eye in an aquatic organism.
They start with a population of organisms, each with a sim-
ple eyespot, a flat patch of light-sensitive tissue sandwiched
between a transparent protective layer and a layer of dark
pigment. They then consider the effect of every possible small
(1%) deformation of the shape of the eyespot on the resolv-
ing power of the eye. They determine which 1% change has
Fish in the genus Poeciliopsis include small the greatest positive effect on the eye’s resolving power and
FIGURE 1.21
minnows like Poeciliopsis occidentalis, shown then repeat the process again and again, deforming the new
here. (Photograph courtesy of David Reznick.) structure by 1% in every possible way at each step. The
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 21
results are shown in Figure 1.22. After 538 changes of 1% each, a simple concave eye Stage
cup evolves; after 1,033 changes of 1%, crude pinhole eyes emerge; after 1,225 changes
1
of 1%, an eye with an elliptical lens is created; and after 1,829 steps, the process finally 176 steps (1%)
comes to a halt because no small changes increase resolving power. The end result is
2
an eye with a spherical gradient lens just like those in fish and other aquatic organisms.
362 steps (1%)
As Nilsson and Pelger point out, 1,829 changes of 1% add up to a substantial amount
of change. For instance, 1,829 changes of 1% would lengthen a 10-cm (4-in.) human 3
finger to a length of 8,000 km (5,500 miles)—about the distance from Los Angeles to
New York and back. Nonetheless, making very conservative assumptions about the 270 steps (1%)
strength of selection, Nilsson and Pelger calculate that this would take only about
364,000 generations. For organisms with short generations, the complete structure of 4
the eye can evolve from a simple eyespot in less than a million years, a brief moment
in evolutionary time. 225 steps (1%)
By comparison, most changes observed in the fossil record are much slower.
Human brain size has roughly doubled in the last 2 million years—a rate of change 5
of 0.00005% per year. This is 10,000 times slower than the rate of change that the
Grants observed in the Galápagos. Moreover, such slow rates of change typify what
can be observed from the fossil record. As we will see, however, the fossil record is 192 steps (1%)
incomplete. It is quite likely that some evolutionary changes in the past were rapid, but
the sparseness of the fossil record prevents us from detecting them. 6
Jenkin published a paper in which he clearly showed that, with blending inheritance,
there could be little or no variation available for selection to act on. The following
example shows why Jenkin’s argument was so compelling. Suppose a population of
one species of Darwin’s finches displays two forms: tall and short. Further suppose that
a biologist controls mating so that every mating is between a tall individual and a short
individual. Then, with blending inheritance, all of the offspring will be the same inter-
mediate height, and their offspring will be the same height as they are. All of the vari-
ation for height in the population will disappear in a single generation. With random
mating, the same thing will occur, though it will take longer. If inheritance were purely
a matter of blending parental traits, then Jenkin would have been right about its effect
on variation. However, as we will see in Chapter 3, genetics accounts for the fact that
offspring are intermediate between their parents and does not assume any kind of
blending.
Another problem arose because selection works by removing variants from pop-
ulations. For example, if finches with small beaks are more likely to die than finches
with larger beaks over the course of many generations, eventually all that will be left
are birds with large beaks. There will be no variation for beak size, and Darwin’s sec-
ond postulate holds that without variation there can be no evolution by natural selec-
tion. For example, suppose the environment changes so that individuals with small
beaks are less likely to die than those with large beaks. The average beak size in the
population will not decrease, because there are no small-beaked individuals. Natural
selection destroys the variation required to create adaptations.
Even worse, as Jenkin also pointed out, there was no explanation of how a pop-
(a)
ulation might evolve beyond its original range of variation. The cumulative evolution
of complex adaptations requires populations to move far outside their original range
of variation. Selection can cull away some characters from a population, but how can
it lead to new types not present in the original population? This apparent contradic-
tion was a serious impediment to explaining the logic of evolution. How could ele-
phants, moles, bats, and whales all descend from an ancient shrewlike insectivore
unless there were a mechanism for creating new variants not present at the beginning?
For that matter, how could all the different breeds of dogs have descended from their
one common ancestor, the wolf (Figure 1.23)?
Remember that Darwin and his contemporaries knew there were two kinds of
variation: continuous and discontinuous. Because Darwin believed that complex
adaptations could arise only through the accumulation of small variations, he thought
discontinuous variants were unimportant. However, many biologists thought that the
discontinuous variants, called “sports” by nineteenth-century animal breeders, were
the key to evolution because they solved the problem of the blending effect. The fol-
lowing hypothetical example illustrates why. Suppose that a population of green birds
has entered a new environment in which red birds are better adapted. Some of Dar-
win’s critics believed that any new variant that was slightly more red would have only
a small advantage and would be rapidly swamped by blending. In contrast, an all-red
(b) bird would have a large enough selective advantage to overcome the effects of blend-
ing and could increase its frequency in the population.
(a) The wolf is
FIGURE 1.23 Darwin’s letters show that these criticisms worried him greatly; and although he
the ancestor of
all domestic dogs, including (b) the tried a variety of counterarguments, he never found one that was satisfactory. The
poodle and the Saint Bernard. These solution to these problems required an understanding of genetics, which was not avail-
transformations were accomplished in able for another half century. As we will see, it was not until well into the twentieth
several thousand generations of artificial century that geneticists came to understand how variation is maintained and Darwin’s
selection. theory of evolution was generally accepted.
01_Adaptation [3p]:How Humans Evolved [5e] 9/26/08 6:10 PM Page 23
FURTHER READING
Browne, J. 1995. Charles Darwin: A Biography, vol. I: Voyaging. New York: Knopf.
Dawkins, R. 1996. The Blind Watchmaker: Why the Evidence of Evolution Reveals
a Universe without Design. New York: Norton.
Dennett, D. C. 1995. Darwin’s Dangerous Idea: Evolution and the Meanings of Life.
New York: Simon & Schuster.
Ridley, M. 1996. Evolution. 2nd ed. Cambridge, Mass.: Blackwell Science.
Weiner, J. 1994. The Beak of the Finch: A Story of Evolution in Our Time. New York: Knopf.
KEY TERMS
adaptations fecundity
natural selection continuous variation
morphology discontinuous variation
equilibrium achondroplasia
stabilizing selection convergence
traits placental mammals
characters marsupials
species blending inheritance
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 24
Chapter 2
Mendelian Genetics
Cell Division and the Role of Chromosomes
in Inheritance
Mitosis and Meiosis
Chromosomes and Mendel’s Experimental
Results
Linkage and Recombination
Molecular Genetics
Genes Are DNA
Some Genes Code for Proteins
Regulatory Sequences Control Gene
Expression
Not All DNA Carries a Message
GENE TICS
MENDELIAN GENETICS
A
lthough none of the main participants in the nineteenth-century debate
about evolution knew it, the key experiments necessary to understand how
genetic inheritance really worked had already been performed by an obscure
Silesian monk, Gregor Mendel, living in what is now Slovakia (Figure 2.1). The son
of peasant farmers, Mendel was recognized by his teachers as an extremely bright stu-
dent, and he enrolled in the University of Vienna to study the natural sciences. While
he was there, Mendel received a first-class education from some of the scientific lu-
minaries of Europe. Unfortunately, Mendel had an extremely nervous disposition:
24
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 25
Mendelian Genetics 25
every time he was faced with an examination, he became physically ill, taking months
to recover. As a result, he was forced to leave the university, after which he joined a
monastery in the city of Brno, more or less because he needed a job. Once there,
Mendel continued to study inheritance, an interest he had developed in Vienna.
Between 1856 and 1863, using the common edible garden pea plant (Figure 2.2),
Mendel isolated a number of traits with only two forms, or variants. For example, one
of the traits he studied was pea color. This trait had two variants: yellow and green.
He also studied pea texture, a trait that also had two variants: wrinkled and smooth. Gregor Mendel,
FIGURE 2.1
Mendel cultivated populations of plants in which these traits bred true, meaning that about 1884, 15
the traits did not change from one generation to the next. For example, crosses (mat- years after abandoning his botanical
experiments.
ings) between plants that bore green peas always produced offspring with green peas,
and crosses between plants that bore yellow peas consistently produced offspring with
yellow peas. Mendel performed a large number of crosses between these different
kinds of true-breeding peas.
Before going further, we need to establish a way to keep track of the results of the
matings. Geneticists refer to the original founding population as the F0 generation,
the offspring of the original founders as the F1 generation, and so on. In this case, the
original true-breeding plants constitute the F0 generation, and the plants created by
crossing true-breeding parents constitute the F1 generation. The offspring of the F1 gen-
eration will be the F2 generation.
In one set of Mendel’s experiments with garden peas, a series of crosses between
green and yellow variants yielded offspring that all bore yellow peas, matching only
one of the parent plants (Figure 2.3). Mendel’s next step was to perform crosses among
the offspring of these crosses. When members of the F1 generation (all of which bore
yellow peas) were crossed, some of the offspring produced yellow seeds and some pro-
duced green seeds. Unlike most of the people who had experimented with plant crosses
before, Mendel performed many of these kinds of crosses and kept careful count of
the numbers of each kind of individual that resulted. These data showed that, in the
F2 generation, there were three individuals with yellow seeds for every one with green
seeds.
Mendel was able to formulate two principles that accounted for his
experimental results.
These two principles account for the pattern of results in Mendel’s breeding experi- Mendel’s genetic
FIGURE 2.2
ments, and, as we shall see, they are the key to understanding how variation is experiments were
preserved. conducted on the common garden pea.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 26
26 CHAPTER 2 • Genetics
Green Yellow
parents parents
F1 generation:
all yellow
F2 generation:
3 yellows
In one of Mendel’s experiments, crossing 1 green
FIGURE 2.3
true-breeding lines of green and yellow
peas led to all yellow offspring. Crossing these F1 individuals led
to an F2 generation with a 3:1 ratio of yellow to green
individuals.
Mendel thought that his findings were important, so he published them in 1866 and
sent a copy of the paper to Karl Wilhelm von Nägeli, a very prominent botanist. Nägeli
was studying inheritance and should have understood the importance of Mendel’s
experiments. Instead, Nägeli dismissed Mendel’s work, perhaps because it contradicted
his own results or because Mendel was an obscure monk. Soon after this, Mendel was
elected abbot of his monastery and was forced to give up his experiments. His ideas
did not resurface until the turn of the twentieth century, when several botanists inde-
pendently replicated Mendel’s experiments and rediscovered the laws of inheritance.
In 1896, the Dutch botanist Hugo de Vries unknowingly repeated Mendel’s exper-
iments with poppies. Instead of publishing his results immediately, however, he cau-
tiously waited until he had replicated his results with more than 30 plant species. Then
in 1900, just as de Vries was ready to send off a manuscript describing his experiments,
a colleague sent him a copy of Mendel’s paper. Poor de Vries; his hot new results were
already 30 years old! About the same time, two other European botanists, Carl Cor-
rens and Erich Tschermak, also duplicated Mendel’s breeding experiments, derived
similar conclusions, and discovered that they, too, had been scooped. Correns and
Tschermak graciously acknowledged Mendel’s primacy in discovering the laws of
inheritance, but de Vries was less magnanimous. He did not cite Mendel in his trea-
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 27
tise on plant genetics, and he refused to sign a petition advocating the construction of
a memorial in Brno commemorating Mendel’s achievements.
When Mendel’s results were rediscovered, they were widely accepted because
scientists now understood the role of chromosomes in the formation of
gametes.
By the time Mendel’s experiments were rediscovered in 1900, it was well known
that virtually all living organisms are built out of cells. Moreover, careful embryolog-
ical work had shown that all the cells in complex organisms arise from a single cell
through the process of cell division. Between the time of Mendel’s initial discovery of
the nature of inheritance and its rediscovery at the turn of the twentieth century, a cru-
cial feature of cellular anatomy was discovered: the chromosome. Chromosomes are
small linear bodies contained in every cell and replicated during cell division (Figure
2.4). Moreover, scientists had also learned that chromosomes are replicated in a spe-
cial kind of cell division that creates gametes. As we will see in subsequent sections,
this research provides a simple material explanation for Mendel’s results. Our current
model of cell division, which was developed in small steps by a number of different
scientists, is summarized in the sections that follow.
Ordinary cell division, called “mitosis,” creates two copies of the chromosomes
present in the nucleus.
When plants and animals grow, their cells divide. Every cell contains within it a body
called the nucleus (plural nuclei; Figure 2.5); when cells divide, their nuclei also divide.
This process of ordinary cell division is called mitosis. As mitosis begins, a cloud of In all plants and
material begins to form in the nucleus, and gradually this cloud condenses into a FIGURE 2.5
animals, every cell
number of linear chromosomes. The chromosomes can be distinguished under the contains a body called the nucleus (the
microscope by their shape and by how they stain. (Stains are dyes added to cells in dark circle in the center of this magni-
the laboratory that allow researchers to distinguish different parts of a cell.) Different fied image). The nucleus contains the
organisms have different numbers of chromosomes, but in diploid organisms, chromosomes.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 28
28 CHAPTER 2 • Genetics
chromosomes come in homologous pairs (pairs whose members have similar shapes
and staining patterns). All primates are diploid, but other organisms have a variety of
arrangements. Diploid organisms also vary in the number of chromosome pairs their
cells have. The fruit fly Drosophila has four pairs of chromosomes, humans have 23,
and some organisms have many more.
Two features of mitosis suggest that the chromosomes play an important role in
determining the properties of organisms. First, the original set of chromosomes is
duplicated so that each new daughter cell has an exact copy of the chromosomes pres-
ent in its parent. This means that, as an organism grows and develops through a
sequence of mitotic divisions, every cell will have the same chromosomes that were
present when the egg and sperm united. Second, the material that makes up the chro-
mosome is present even when cells are not dividing. Cells spend little of their time
dividing; most of the time they are in a “resting” period, doing what they are supposed
to do as liver cells, muscle cells, bone cells, and so on. During the resting period, chro-
mosomes are not visible. However, the material that makes up the chromosomes is
always present in the cell.
In meiosis, the special cell division process that produces gametes, only half of
the chromosomes are transmitted from the parent cell to the gamete.
The sequence of events that occur during mitosis is quite different from the
sequence of events during meiosis, the special form of cell division leading to the pro-
duction of gametes. The key feature of meiosis is that each gamete contains only one
copy of each chromosome, whereas cells that undergo mitosis contain a pair of homol-
ogous chromosomes. Cells that contain only one copy of each chromosome are said
to be haploid (Figure 2.6). When a new individual is conceived, a haploid sperm from
the father unites with a haploid egg from the mother to produce a diploid zygote. The
zygote is a single cell that then divides mitotically over and over to produce the mil-
lions and millions of cells that make up an individual’s body.
Mendel’s two principles can be deduced from the assumption that genes are
carried on chromosomes.
In 1902, less than two years after the rediscovery of Mendel’s findings, Walter Sutton,
a young graduate student at Columbia University, made the connection between chro-
mosomes and the properties of inheritance revealed by Mendel’s principles. Recall that
the first of Mendel’s two principles states that an organism’s observed characteristics
are determined by particles acquired from each of the parents. This concept fits with
the idea that genes reside on chromosomes, because individuals inherit one copy of
each chromosome from each parent. The idea that observed characteristics are deter-
mined by genes from both parents is consistent with the observation that mitosis trans-
mits a copy of both chromosomes to every daughter cell, so every cell contains copies
of both the maternal and the paternal chromosomes. Mendel’s second principle states
that genes segregate independently. The observation that meiosis involves the creation
of gametes with only one of the two possible chromosomes from each homologous
pair is consistent with two notions: (1) that one gene is inherited from each parent, and
(2) that each of these genes is equally likely to be transmitted to gametes. Not every-
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 29
Diploid parent cell Diploid daughter cells Diploid parent cell Haploid gametes
Diploid cells have
FIGURE 2.6
n pairs of homolo-
gous chromosomes; n varies widely
among species, but here n = 1. Mem-
bers of homologous pairs may differ in
the alleles they carry at specific sites
along the chromosome. (a) Mitosis
duplicates the chromosomes. (b) Meio-
(a) (b) sis creates gametes that carry only one
member of each homologous pair of
Mitosis Meiosis chromosomes.
one agreed with Sutton, but over the next 15 years, T. H. Morgan and his colleagues
at Columbia performed many experiments that proved Sutton right.
Different varieties of a particular gene are called “alleles.” Individuals with two
copies of the same allele are “homozygous”; individuals with different alleles
are “heterozygous.”
To see more clearly the connection between chromosomes and the results of
Mendel’s experiments with the peas, we need to introduce some new terms. The word
gene is used to refer to the material particles carried on chromosomes. Later you will
learn that genes are made of a molecule called “DNA.” Alleles are varieties of genes
with the same effect on the organism. Individuals with two copies of the same allele
are homozygous for that allele and are called “homozygotes.” When individuals carry
copies of two different alleles, they are said to be heterozygous for those alleles and
are called “heterozygotes.”
Consider the case in which all the yellow individuals in the parental generation
carry two genes for yellow pea color, one on each chromosome. We will use the sym-
bol A for this allele. Thus, these plants are homozygous (AA) for yellow pea color. All
of the individual plants with green peas are homozygous for a different allele, which
we denote a, so green-pea plants are aa. As we will see, this is the only pattern that is
consistent with Mendel’s model. What happens if we cross two yellow-pea plants with
each other, or two green-pea plants with each other? Because they are homozygous,
all of the gametes produced by the yellow parents will carry the A allele. This means
that all of the offspring produced by the crossing of two AA parents will also be
homozygous for that A allele and therefore will also produce yellow peas. Similarly,
all of the gametes produced by parents that are homozygous for the a allele will carry
the a allele; when the gametes of two aa parents unite, they will produce only aa indi-
viduals with green peas. Thus, we can explain why each type breeds true.
Next let’s consider the offspring of a mating between a true-breeding green par-
ent and a true-breeding yellow parent (Figure 2.7). The green parent produces only
a gametes, and the yellow parent produces only A gametes. Thus, every one of their
offspring inherits an a gamete from one parent and an A gamete from the other par-
ent. According to Mendel’s model, all the individuals in the F1 generation will be Aa.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 30
30 CHAPTER 2 • Genetics
A A a a Parents
A A A A a a a a
Gametes
A A A A a a a
a
Given that Mendel discovered that all offspring of such crosses have yellow peas, it
must be that heterozygotes bear yellow peas. To describe these effects, geneticists use
the following four terms:
As you can see in Table 2.1, AA and Aa individuals have the same phenotype but dif-
ferent genotypes; knowing an individual’s observable characteristics, or phenotype,
TABLE 2.1 The relation- does not necessarily tell you its genetic composition, or genotype.
ship between
genotype and phenotype in Mendel’s
A cross between heterozygous parents produces a predictable mixture of all
experiment on pea color.
three genotypes.
Genotype Phenotype
Now consider the second stage of Mendel’s experiment: crossing members of the
AA Yellow
F1 generation with each other to create an F2 generation (Figure 2.8). We have seen
Aa Yellow that every individual in the F1 generation is heterozygous, Aa. This means that half of
aa Green their gametes contain a chromosome with an A allele and half with an a allele.
(Remember, meiosis produces haploid gametes.) On average, if we draw pairs of
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 31
A a A a F1 parents
A a a A A a a A
Gametes from
F1 parents
a A A a a A A a
Crosses of the F1
FIGURE 2.8
heterozygotes
yielded a 3:1 ratio of phenotypes in
Mendel’s experiments. All the parents
are heterozygous, or Aa. This means
A a that half of the gametes produced by
each parent carry the A allele, and the
A A a a Offspring other half produce the a allele. Thus, 1⁄4
(F2)
of the zygotes will be AA, 1⁄2 will be
Aa, and the remaining 1⁄4 will be aa.
A a Because A is dominant, 3⁄4 of the off-
spring (1⁄4 AA + 1⁄2 Aa) will produce
yellow seeds.
gametes at random from the gametes produced by the F1 generation, 1⁄4 of the indi-
viduals will be AA, 1⁄2 will be Aa, and 1⁄4 will be aa.
To see why there is a 1:2:1 ratio in the F2 generation, it is helpful to construct an
event tree, like the one shown in Figure 2.9. First pick the paternal gamete. Every male
in the F1 generation is heterozygous: he has one chromosome with an A allele and one
with an a allele. Thus there is a probability of 1⁄2 of getting a sperm that carries an A,
and a probability of 1⁄2 of getting a sperm that carries an a. Suppose you select an A
by chance. Now pick the maternal gamete. Once again you have a 1⁄2 chance of get-
ting an A and a 1⁄2 chance of getting an a. The probability of getting two As, one from
the father and one from the mother, is
Pr(AA) = Pr(A from dad) ✕ Pr(A from mom) = 1⁄2 ✕ 1⁄2 = 1⁄4
If you repeated this process a large number of times—first picking male gametes
and then picking female gametes—roughly 1⁄4 of the F2 individuals produced would
be AA. Similar reasoning shows that 1⁄4 would be aa. One-half would be Aa because
there are two ways of combining the A and a alleles: the A could come from the
mother and the a from the father, or vice versa. Because both AA and Aa individuals
have yellow peas, there will be three yellow individuals for every green one among the
offspring of the F1 parents. Another way to visualize this result is to construct a dia-
gram called a Punnett square, like the one shown in Figure 2.10.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 32
32 CHAPTER 2 • Genetics
Mendel also performed experiments that involved two characters. For example, he
crossed individuals that bred true for two different traits: pea color and seed texture.
He crossed plants with smooth, yellow seeds and plants with wrinkled, green seeds.
All of the F1 individuals were smooth and yellow, but the F2 individuals occurred in
the following ratio:
Female gametes
1 1
2 2
A A A A A A a a a a a a
A
A A 1 A 5 AA A 1 a 5 aA
A
1 1 1 1 1
Pr(AA) 5 Pr(aA) 5
2 A 2 3 2 2 3 2 This diagram, called a Punnett square, pro-
FIGURE 2.10
A 1 1 vides another way to see why there is a
5
Male gametes
4 5
4 1:2:1 genotypic ratio among offspring in the F2 generation. The
A
horizontal axis is divided in half to reflect the equal proportion of A
a and a eggs. The vertical axis is divided according to the proportion
a a 1 A 5 Aa a 1 a 5 aa of sperm carrying each allele, and again it is divided in half. The
areas of squares formed by the intersection of the vertical and hori-
1 a 1 1 1 1
Pr(aA) 5 Pr(aa) 5 zontal dividing lines give the proportion of zygotes that result from
2
a 2 3 2 2 3 2
each of the four possible fertilization events: AA = 0.25 (one of
a 1 1 the four squares), Aa = 0.25 + 0.25 = 0.50 (two of the four
5 5
4 4 squares), and aa = 0.25 (one of the four squares). Thus, zygotes
a
will have genotypes in the ratio 1:2:1.
whether it will have B or b; it is equally likely to carry either of these alleles. Mendel’s
experiments convinced him that all traits segregate independently. Today, however, we
know that independent segregation occurs only when the traits measured are con-
trolled by genes that reside on different chromosomes.
It turns out that the genes for a particular character occur at a particular site on
a particular chromosome. Such a site is called a locus (plural loci). Loci are arranged
on the chromosomes in a line, like beads on a string. A particular locus may hold any
of several alleles. The gene for seed color is always in the same position on a particu-
lar chromosome, whether it codes for green or for yellow seeds. Keep in mind that
because organisms have more than one pair of chromosomes, the locus for seed color
may be located on one chromosome and the locus for seed texture located on another
chromosome. All of the genes carried on all of the chromosomes are referred to as the
genome.
Traits may not segregate independently if they are affected by genes on the
same chromosome.
Mendel’s conclusion that traits segregate independently is true only if the loci that
affect the traits are on different chromosomes, because links between loci on the same
chromosome alter the patterns of segregation. When loci for different traits occur on
the same chromosome, they are said to be linked; loci on different chromosomes are
said to be unlinked. You might think that genes at two loci on the same chromosome
would always segregate together as if they were a single gene. If this were true, then
a gamete receiving a particular chromosome from one of its parents would get all of
the same genes that occurred on that chromosome in its parent. There would be no
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 34
Box 2.1
More on Recombination
The first step to a deeper understanding of recombination is of squares that contain each phenotype. There are nine
to see the connections among Mendel’s two-trait experi- smooth-yellow squares, three wrinkled-yellow squares,
ment, independent segregation, and chromosomes. Mendel three smooth-green squares, and one wrinkled-green square.
crossed a smooth-yellow (AABB) parent and a wrinkled- Thus the 9:3:3:1 ratio of phenotypes that Mendel observed
green (aabb) parent to produce members of an F1 genera- was consistent with the assumption that genes on different
tion. Smooth-yellow parents produce only AB gametes, and chromosomes segregate independently. If the genes control-
wrinkled-green parents produce only ab gametes; all mem- ling these traits did not segregate independently, the ratio of
bers of the F1 generation are, therefore, AaBb. If we assume phenotypes would be different, as we will see next.
that the genes for seed color and seed texture enter gametes During meiosis, chromosomes frequently become dam-
independently, then each of the four possible types—AB, aged, break, and recombine. This process, called “crossing
Ab, aB, and ab—will be represented by 1⁄4 of the gametes over,” creates chromosomes with combinations of genes not
produced by members of the F1 generation (Figure 2.11). present in the parent (see text). In the preceding example,
With this information, we can construct a Punnett remember that all members of the F1 generation were AaBb.
square predicting the proportions of each genotype in the F2 Now suppose that the locus controlling seed color and the
generation (Figure 2.12). Once again, we divide the vertical locus controlling seed texture are carried on the same chro-
and horizontal axes in proportion to the frequency of each mosome. Further assume that when the chromosomes are
type of gamete, in this case dividing each axis into four duplicated during meiosis, a fraction r of the time there is
equal parts. The areas of the rectangles formed by the inter- crossing over (Figure 2.13), and a fraction 1 – r of the time
section of the vertical and horizontal dividing lines give the there is no crossing over. Next, chromosomes segregate
proportions of zygotes that result from each of the 16 pos- independently into gametes. The types of chromosomes
sible fertilization events. Because the area of each cell in this present in the parental generation, Ab and ab, each occur in
matrix is the same, we can determine the phenotypic ratios a fraction (1 – r)/2 of the gametes; the novel, recombinant
of zygotes that result from each of the 16 possible fertiliza- types Ab and aB each occur in a fraction r/2 of the gametes
tion events in this example by simply counting the number (Figure 2.14).
Resulting Probability
gamete of gamete
34
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 35
Female gametes
1 1 1 1
4 4 4 4
AB Ab aB ab
1 AA AA Aa Aa
4 AB BB Bb BB Bb
1 AA AA Aa Aa
Male gametes
4 Ab Bb bb Bb bb
This Punnett square shows why there is a
FIGURE 2.12
9:3:3:1 phenotypic ratio among offspring of
1 Aa Aa aa aa the F2 generation when the genes for the two traits, seed color and
4 aB BB Bb BB Bb
seed texture, are carried on different chromosomes. The rings in each
square show the color (green or yellow) and seed texture (wrinkled or
smooth) of the phenotype associated with each genotype.
1 Aa Aa aa aa
4 ab Bb bb Bb bb
a
1 (1 – r)
1 2
2 b
A a
B b
A B 1
2 A
1– r 1 (1 – r)
2
a b B
A a
(a)
B b
A b A
1r
r 2
a B 1 b
2
a A
(b)
B b
A b 1 a
2 1r
2
a B B
35
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 36
Female gametes
1–r r r 1–r
2 2 2 2
AB AB AB AB AB Ab Ab Ab aB aB aB ab ab ab ab ab
AB
This Punnett AB AA AA Aa Aa
FIGURE 2.15 BB Bb BB Bb
square shows 1–r AB
how to calculate the frequency of each 2 AB
phenotype among offspring in the F2 (1 – r)2 r (1 – r) r (1 – r) (1 – r)2
AB
generation if the genes for seed color 4 4 4 4
and seed texture are carried on the AB
same chromosome. The horizontal axis Ab AA AA Aa Aa
is divided according to the proportion r Bb bb Bb bb
of each type of egg: AB, Ab, aB, and Ab
2
r (1 – r) r2 r2 r (1 – r)
Male gametes
Now we can use a Punnett square to calculate the fre- each genotype. If recombination rates are low, most mem-
quency of each of the 16 possible genotypes in the F2 gen- bers of the F2 generation will be of three genotypes—
eration (Figure 2.15). As before, we divide the vertical and AABB, AaBb, and aabb—just as if there were only two alle-
horizontal axes in proportion to the relative frequency of les, AB and ab. If the recombination rate is higher, however,
each type of gamete, and the area of the rectangles formed more of the novel recombinant genotypes will be produced.
by the intersection of these grid lines gives the frequency of
recombination. However, chromosomes frequently tangle and break as they are repli-
cated during meiosis. Thus, chromosomes are not always preserved intact, and genes
on one chromosome are sometimes shifted from one member of a homologous pair
to the other (see Figure 2.13). We call this process crossing over. Linkage reduces the
rate of recombination but does not eliminate it altogether. The rate at which recom-
bination generates novel combinations of genes at two loci on the same chromosome
depends on the likelihood that a crossing-over event will occur. If two loci are located
close together on the chromosome, crossing over will be rare and the rate of recom-
bination will be low. If the two loci are located far apart, then crossing over will be
common and the rate of recombination will approach the rate for genes on different
chromosomes. This process is discussed more fully in Box 2.1.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 37
Molecular Genetics 37
MOLECULAR GENETICS
Genes are segments of a long molecule called DNA, which is contained in
chromosomes.
In the first half of the twentieth century, biologists learned a lot about the cellular
events that take place during meiosis and mitosis, and began to understand the chem-
istry of reproduction. For instance, by 1950 it was known that chromosomes contain
two structurally complex molecules: protein and deoxyribonucleic acid, or DNA. It
had also been determined that the particle of heredity postulated by Mendel was DNA,
not protein, though exactly how DNA might contain and convey the information
essential to life was still a mystery. Then in 1953, two young biologists at Cambridge
University, Francis Crick and James Watson, made a discovery that revolutionized biol-
ogy: they deduced the structure of DNA. Watson and Crick’s elucidation of the struc-
ture of DNA was the wellspring of a great flood of research that continues to provide
a deep and powerful understanding of how life works at the molecular level. We now
know how DNA stores information and how this information controls the chemistry
of life, and this knowledge explains why heredity leads to the patterns Mendel
described in pea plants, and why there are sometimes new variations.
Understanding the chemical nature of the gene is critical to the study of human
evolution: (1) molecular genetics links biology to chemistry and physics, and
(2) molecular methods help us reconstruct the evolutionary history of the
human lineage.
Modern molecular genetics, the product of Watson and Crick’s discovery, is a field
of great intellectual excitement. Every year yields new discoveries about how living
things work at the molecular level. This knowledge is deeply important because it links
biology to chemistry and physics. One of the grandest goals of science is to provide a
single consistent explanatory framework for the way the world works. We want to
place evolution in this grand scheme of scientific explanation. It is important to be able
to explain not only how new species of plants and animals arise, but also how a wide
range of phenomena—from the origin of stars and galaxies to the rise of complex
societies—have evolved. Modern molecular biology is profoundly important because
it connects physical and geochemical evolution to Darwinian processes.
Molecular genetics also provides data that help biologists and anthropologists
reconstruct evolutionary history. As we will see in Chapter 4, comparing the DNA
sequences of different species allows us to reconstruct their evolutionary histories. For
example, this kind of analysis tells us that humans share a more recent common ances-
tor with chimpanzees than members of either species share with gorillas. The same
data tell us that the last common ancestor of chimpanzees and humans lived between
5 and 7 mya. Patterns of variation in DNA sequences within species are also inform-
ative. In Chapter 13, we will see that patterns of genetic variation within the human
species allow anthropologists to figure out when the first modern humans left Africa
and where they went. In Chapter 14, we will see that the patterns of genetic variation
also provide important clues about how natural selection has shaped adaptation within
the human species.
In this section, we provide a very brief and highly selective introduction to
molecular genetics. Our aim is to give enough background to allow students to under-
stand the molecular evidence about human evolution. Students who want a richer
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 38
38 CHAPTER 2 • Genetics
understanding of the science of molecular genetics and its important implications for
human societies should consult the Further Readings section at the end of this chapter.
The discovery of the structure of DNA was fundamental to genetics because the struc-
ture itself implied how inheritance must work. Each chromosome contains a single
DNA molecule roughly 2 m (about 6 ft) long that is folded up to fit in the nucleus.
DNA molecules consist of two long strands, and each strand has a “backbone” of
alternating sequences of phosphate and sugar molecules. Attached to each sugar is one
of four molecules, collectively called bases: adenine, guanine, cytosine, or thymine. The
two strands of DNA are held together by very weak chemical bonds, called “hydro-
gen bonds,” which connect some of the bases on different strands. Thymine bonds
only with adenine, and guanine bonds only with cytosine (Figure 2.16).
The repeating four-base structure of DNA allows the molecule to assume a vast
number of distinct forms. Each DNA configuration is exactly like a message written
in an alphabet with letters that stand for each of the four bases (T for thymine, A for
adenine, G for guanine, and C for cytosine). Thus,
TCGGTAGTAGTTACGG
ATCCGGATGCAATCCA
is another message. Because the DNA in a single chromosome is millions of bases long,
there is room for a nearly infinite variety of messages.
These messages would be of no consequence if they were not preserved over time
and transmitted faithfully; DNA is uniquely suited to this task. Each of the stagger-
ing number of DNA molecules that could exist in nature are equally stable chemically.
DNA is not the only complex molecule with many alternative forms, but other mol-
ecules have some forms that are less stable than others. Such molecules would be
unsuitable for carrying information because the messages would become garbled as the
molecules changed toward a more stable form. What makes DNA unusual is that all
of its nearly infinite number of forms are equally stable.
In addition to preserving a message faithfully, hereditary material must be replic-
able. Without the ability to make copies of itself, the genetic message could not be
spread to offspring, and natural selection would be impossible. DNA is replicated
within cells by a highly efficient cellular machinery: it first unzips the two strands, and
then, with the help of other specialized molecular machinery, it adds complementary
bases to each of the strands until two identical sugar and phosphate backbones are
built (Figure 2.17). There are also mechanisms that “proofread” the copies and cor-
rect errors that crop up. These proofreading mechanisms are astoundingly good; they
miss only one error in every billion bases replicated.
For DNA to play an important role in evolution, different DNA messages must
lead to different phenotypes. Here the story gets more complicated because DNA
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 39
Molecular Genetics 39
T A S
S P
P
Base C S
S G
Phosphate (P)
Deoxyribose P
P
sugar (S)
C G S
S
Base pair
P P
A T 5′
S S
3′
The chemical structure of
FIGURE 2.16
P P DNA consists of two long
5′ backbones made of alternating sugar and phosphate
T 3′
A S molecules. One of four bases—adenine (A), guanine
S (G), cytosine (C), or thymine (T)—is attached to
each of the sugars. The two strands are connected to
each other by hydrogen bonds (dotted lines) between
certain pairs of bases. Thymine bonds only to adenine,
and guanine bonds only to cytosine.
affects phenotypes in a number of different ways. The three most important ways that
DNA affects phenotypes are the following:
1. DNA in protein coding genes (or protein coding sequences) specifies the struc-
ture of proteins. Proteins play many important roles in the machinery of life. In
particular, many proteins are enzymes, which regulate much of the biochemical
machinery of organisms.
2. DNA in regulatory genes (or regulatory sequences) determines the conditions
under which the message encoded in a protein coding gene will be expressed.
Regulatory genes play a crucial role in shaping the differentiation of cells dur-
ing development.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 40
40 CHAPTER 2 • Genetics
C G
G C
T A
C G
T A
G C
C G
Parent DNA
A T
T A
G C
A T
C G
T A
G C
C G Growing
ends of new
DNA strands
T A A T
C G C G
G C C G
T A T A
G C A T
A T T A
When DNA is T A C G
FIGURE 2.17
replicated, the two
strands of the parent DNA are separated
and two daughter DNA strands are formed. Daughter DNAs
3. Other DNA specifies the structure of several different kinds of ribonucleic acid
(RNA) molecules which perform important cellular functions. These include
RNA molecules that play a crucial role in the machinery of protein synthesis and
other RNA molecules that help regulate the expression of other genes.
The cells and organs of living things are made up a very large number of chemical com-
pounds, and it is this combination of compounds that gives each organism its char-
acteristic form and structure. All organisms use the same raw materials, but they
achieve different end results. How does this happen?
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 41
Molecular Genetics 41
Product 5
Substrate
Enzyme X catalyzes Product 2 Enzyme Y catalyzes
reaction that converts reaction that converts
substrate to product 2. Product 1 product 2 to product 5.
The answer is that enzymes present in the cell determine what the raw materials
are transformed into when cells are built (Figure 2.18). The best way to understand
how enzymes determine the characteristics of organisms is to think of an organism’s
biochemical machinery as a branching tree. Enzymes act as switches to determine what
will happen at each node and thus what chemicals will be present in the cell. For exam-
ple, glucose serves as a food source for many cells, meaning that it provides energy and
a source of raw materials for the construction of cellular structures. Glucose might ini-
tially undergo any one of an extremely large number of slow-moving reactions. The
presence of particular catalytic enzymes will determine which reactions occur rapidly
enough to alter the chemistry of the cell. For example, some enzymes lead to the metab-
olism of glucose and the release of its stored energy. At the end of the first branch, there
is another node representing all of the reactions that could involve the product(s) of
the first branch. Again, one or more enzymes will determine what happens next.
This picture has been greatly simplified. Real organisms take in many different
kinds of compounds, and each compound is involved in a complicated tangle of
branches that biochemists call “pathways.” Real biochemical pathways are very com-
plex. One set of enzymes causes glucose to be shunted to a pathway that yields energy.
A different set of enzymes causes the glucose to be shunted to a pathway that binds
glucose molecules together to form glycogen, a starch that functions as energy stor-
age. The presence of a different set of enzymes would lead to the synthesis of cellulose,
a complex molecule that provides the structural material in plants. Enzymes play roles
in virtually all cellular processes—from the replication of DNA and division of cells
to the contraction and movement of muscles.
Some proteins have crucial structural functions in living things. For example, your
hair is made up mainly of a protein called “keratin,” and your ligaments and tendons
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 42
42 CHAPTER 2 • Genetics
are strengthened by another protein, called “collagen.” Other proteins act as tiny
mechanical contraptions that accomplish many important functions. Some are tiny
valves for regulating what goes into and out of cells. Still others, like insulin, convey
chemical signals from one part of the body to another or act as receptors that respond
to these messages.
Proteins are constructed of amino acids. There are 20 different amino acid mol-
ecules. All amino acids have the same chemical backbone, but they differ in the chem-
ical composition of the side chain connected to this backbone (Figure 2.19). The
sequence of amino acid side chains, called the primary structure of the protein, is what
makes one protein different from others. You can think of a protein as a very long rail-
road train in which there are 20 different kinds of cars, each representing a different
amino acid. The primary structure is a list of the types of cars in the order that they
occur.
When proteins are actually doing their business, they are folded in complex ways.
The three-dimensional shape of the folded protein, called the tertiary structure, is cru-
cial to its catalytic function. The way the protein folds depends on the sequence of
amino acid molecules that make up its primary sequence. This means that the func-
tion of enzymes depends on the sequence of the amino acids that make them up. (Pro-
teins also have secondary structure and sometimes quaternary structure; but to keep
things simple, we will ignore those levels here.)
These ideas are illustrated in Figure 2.20, which shows the folded shape of part
of a hemoglobin molecule, a protein that transports oxygen from the lungs to the tis-
sues via red blood cells. As you can see, the protein folds into a roughly spherical glob,
and oxygen is bound to the protein near the center of the glob. Sickle-cell anemia, a
condition common among people in West Africa and among African Americans, is
caused by a single change in the primary sequence of amino acids in the hemoglobin
molecule. Glutamic acid is the sixth amino acid in normal hemoglobin molecules, but
in people afflicted with sickle-cell anemia, valine is substituted for glutamic acid. This
single substitution changes the way that the molecule folds, and reduces its ability to
bind oxygen.
Now we return to our original question: how does the information contained in
DNA—its sequence of bases—determine the structure of proteins? Remember that
DNA encodes messages in a four-letter alphabet. Researchers determined that these let-
ters are combined into three-letter “words” called codons, each of which specifies a
particular amino acid. Because there are four bases, there are 64 possible three-letter
combinations for codons (4 possibilities for the first base times 4 for the second times
4 for the third, or 4 ✕ 4 ✕ 4 = 64). Of these codons, 61 are used to code for the 20
amino acids that make up proteins. For example, the codons GCT, GCC, GCA, and
GCG all code for alanine; GAT and GAC code for asparagine; and so on. The remain-
ing three codons are “punctuation marks” that mean either “start, this is the begin-
ning of the protein” or “stop, this is the end of the protein.” Thus, if you can identify
the base pairs, it is a simple matter to determine what proteins are encoded on the
DNA.
You might be wondering why several different codons code for the same amino
acid. This redundancy serves an important function. Because a number of processes
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 43
NH2
+
C NH2
NH
CH2 –
O NH2 O O
CH2 C C SH
CH3 CH2 CH2 CH2 CH2
H2N CH C OH H2N CH C OH H2 N CH C OH H2 N CH C OH H2 N CH C OH
O O O O O
Alanine (Ala) Arginine (Arg) Asparagine (Asn) Aspartic acid (Asp) Cysteine (Cys)
–
O NH2 O O
C C CH3
HN +
NH
CH2 CH2 CH2
CH2 CH2 H CH2 CH3 CH
H2N CH C OH H2N CH C OH H 2N CH C OH H2 N CH C OH H2 N CH C OH
O O O O O
Glutamine (Gln) Glutamic acid (Glu) Glycine (Gly) Histidine (His) Isoleucine (Ile)
+
NH3
CH2 CH3
CH2 S
CH3 CH3
CH CH2 CH2 CH2
CH2 CH2 CH2 CH2 CH2 CH2
H2N CH C OH H2N CH C OH H2 N CH C OH H2 N CH C OH H2 N CH C OH
O O O O O
Leucine (Leu) Lysine (Lys) Methionine (Met) Phenylalanine (Phe) Proline (Pro)
H OH
N
All amino acids
OH CH3 CH3 CH3 FIGURE 2.19
share the same
CH2 HCOH CH2 CH2 CH
chemical backbone, here colored blue.
H2N CH C OH H2N CH C OH H2 N CH C OH H2 N CH C OH H2 N CH C OH They differ according to the chemical
O O O O O structure of the side group attached to
Serine (Ser) Threonine (Thr) Tryptophan (Trp) Tyrosine (Tyr) Valine (Val) the backbone, here colored pink.
Folded ß-globin
polypeptide
44 CHAPTER 2 • Genetics
can damage DNA and cause one base to be substituted for another, redundancy
decreases the chance that a random change will alter the primary sequence of the pro-
tein produced. Proteins represent complex adaptations, so we would expect most
changes to be deleterious (harmful). But because the code is redundant, many substi-
tutions have no effect on the message of a particular stretch of DNA. The importance
of this redundancy is underscored by the fact that the most common amino acids are
the ones with the greatest number of codon variants.
Before DNA is translated into proteins, its message is first transcribed into
messenger RNA.
DNA can be thought of as a set of instructions for building proteins, but the real
work of synthesizing proteins is performed by other molecules. The first step in the
translation of DNA into protein occurs when a facsimile of one of the strands of DNA,
which will serve as a messenger or chemical intermediary, is made, usually in the cell’s
nucleus. This copy is ribonucleic acid, or RNA. RNA is similar to DNA, except that
it has a slightly different chemical backbone, and the base uracil (denoted U) is sub-
stituted for thymine. RNA comes in several forms, many of which aid in protein syn-
thesis. The form of RNA used in this first step is messenger RNA (mRNA).
The ribosome then synthesizes a particular protein by reading the mRNA copy
of the gene.
Meanwhile, amino acid molecules are bound to a different kind of RNA, called
transfer RNA (tRNA). Each tRNA molecule has a triplet of bases, called an anticodon,
at a particular site (Figure 2.21). Each type of tRNA is bound to the amino acid whose
codon binds to the anticodon on the tRNA. For example, one of the codons for the
amino acid alanine is the base sequence GCU, and GCU binds only to the anticodon
CGA. Thus the tRNA with the anticodon CGA binds only to the amino acid alanine.
The next step in the process involves the ribosomes. Ribosomes are small cellu-
lar organelles. Composed of protein and RNA, organelles are cellular components that
perform a particular function, analogous to the way organs such as the liver perform
a function for the body as a whole. The mRNA first binds to ribosomes at a binding
site and then moves through the binding site one codon at a time. As each codon of
mRNA enters the binding site, a tRNA with a complementary anticodon is drawn
from the complex soup of chemicals inside the cell and bound to the mRNA. The
amino acid bound to the other end of the tRNA is then detached from the tRNA and
added to one end of the growing protein chain. The process repeats for each codon,
continuing until the end of the mRNA molecule passes through the ribosome. Voilà!
A new protein is ready for action.
So far, our description of protein synthesis applies to almost all organisms. How-
ever, most of this information was learned through the study of Escherichia coli, a bac-
terium that lives in the human gut. Like other bacteria, E. coli belongs to a group of
organisms called the prokaryotes because it does not have a chromosome or a cell
nucleus. In prokaryotes, the DNA sequence that codes for a particular protein is unin-
terrupted. A stretch of DNA is copied to RNA and then translated into a protein. For
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 45
Molecular Genetics 45
DNA template
T A C T C A T T T A C C C G T A T G A C G G AG GG T
A T G A G T A A AT G GGC A T A C T G C C T C C C A
AU G AG U A A AUG GG C AA U C U G C C U C C C A
mRNA transcript
Aminoacyl-
Nucleus
tRNA
synthetase
adding amino
Cytoplasm acid cysteine
to right tRNA
Cys
Growing protein
Met
Ser Ribosome
Lysine tRNA Lys Tyr
ACG
without amino
Trp
acid lysine Ala
Tyrosine tRNA
with amino
acid tyrosine
UUU AUG
AGCCGU
mRNA AUGAGUAAAUCGGCAUACUGCCUCCCA
Ribosome moving
down mRNA
many years, biologists thought that the same would be true of eukaryotes (organisms
like plants, birds, and humans that have chromosomes and a cell nucleus).
Beginning in the 1970s, new recombinant DNA technology allowed molecular
geneticists to study eukaryotes. These studies revealed that in eukaryotes, the segment
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 46
46 CHAPTER 2 • Genetics
of DNA that codes for a protein is almost always interrupted by at least one—and
sometimes many—noncoding sequences called introns. (The protein coding sequences
are called exons.) Protein synthesis in eukaryotes includes one additional step not men-
tioned in our discussion so far: after the entire DNA sequence is copied to make an
mRNA molecule in the nucleus, the intron-based parts of the mRNA are snipped out
and the mRNA molecule is spliced back together. Then, the mRNA is exported out of
the nucleus, and protein synthesis takes place.
Alternative splicing allows the same DNA sequence to code for more than
one protein.
When the introns are snipped out and the RNA is spliced back together, not all
of the exons are necessarily included. This means that the same sequence of DNA can
yield many different mRNAs and code for many different proteins. Figure 2.22
shows how this can work. A hypothetical protein has four exons (in colors) and three
introns (in gray). The entire DNA sequence is transcribed into RNA, and the introns
are removed. The exons are always spliced back together in their original order, but
not all of the exons are included in every new RNA molecule. By including different
exons in the final RNA, different proteins can be produced.
Alternative splicing seems to be important. It has been estimated that more than
half of the protein coding sequences in the human genome yield more than one pro-
tein, and some yield as many as 10. Some biologists think that this flexibility is impor-
tant in multicellular organisms in which different types of proteins are required for the
function of diverse cell types. Adherents of this hypothesis think that introns prolif-
erated at the origin of eukaryotes. Other biologists think that introns are weakly dele-
terious bits of DNA that are maintained in eukaryotes because their population sizes
are much smaller than prokaryotes. As we will see in Chapter 3, small populations are
subject to a random nonadaptive evolutionary process called genetic drift that usually
works against natural selection. In bacterial populations numbering in the hundreds
of millions, drift is weak and selection eliminates the introns. In the much smaller
eukaryotic populations, drift is strong enough to maintain introns.
The DNA sequence in regulatory genes determines when protein coding genes
are expressed.
Gene regulation in the bacterium Escherichia coli provides a good example of how the
DNA sequence in regulatory genes controls gene expression. E. coli uses the sugar glu-
cose as its primary source of energy. When glucose runs short, E. coli can switch to
other sugars, such as lactose, but this switch requires a number of enzymes that allow
lactose to be metabolized. The genes for these lactose-specific enzymes are always pres-
ent, but they are not expressed when there is plenty of glucose available. This is effi-
cient because it would be wasteful to produce these enzymes if they were not needed.
The genes for making enzymes that allow the metabolism of lactose to glucose are
expressed only when glucose is in short supply and sufficient lactose is present. Two
regulatory sequences are located near the protein coding genes that encode the amino
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 47
Molecular Genetics 47
DNA pre-mRNA
coding sequence primary transcript Translation
Alternative
Transcription splicings Translation
Translation
A hypothetical protein has four exons (red, green, blue, and brown) and three
FIGURE 2.22
introns (gray). The entire DNA sequence is transcribed into a pre-mRNA
molecule. After introns are snipped out, the exons are spliced back together to yield three different
mRNAs, and these are translated into three different proteins.
acid sequence for three enzymes necessary for lactose metabolism. When there is glu-
cose in the environment, a repressor protein binds to one of the two regulatory
sequences, thereby preventing the protein coding genes from being transcribed (Fig-
ure 2.23a). When glucose is absent, the repressor protein changes shape and does not
bind to the DNA in the regulatory sequence (Figure 2.23b). The second regulatory
sequence is an activator. In the presence of lactose, an activator protein binds to
this DNA sequence, greatly increasing the rate at which the protein coding genes are
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 48
48 CHAPTER 2 • Genetics
(a)
Glucose present
no transcription
repressor
(b)
RNA polymerase
No glucose
No lactose
basal level
of transcription
No glucose
(c)
Lactose present
RNA polymerase
activator activated level
of transcription
This diagram shows how regulatory sequences binding activator and repressor
FIGURE 2.23
proteins control the expression of the enzymes necessary to digest lactose in
E. coli. (a) When glucose is present, the repressor protein binds to the regulatory sequence, thus pre-
venting the enzyme RNA polymerase from creating an mRNA from the DNA template. (b) When glu-
cose is absent and there is no lactose that the RNA polymerase can bind, the genes necessary for
metabolism of lactose are expressed but at a low (basal) level. (c) When lactose is present, the activa-
tor protein can bind to its regulatory sequence, thereby greatly increasing the rate at which the RNA
polymerase binds and increasing the level of gene expression.
transcribed (Figure 2.23c). The specific DNA sequences of the regulatory genes con-
trol whether or not the repressor and activator proteins that control DNA transcrip-
tion bind to the DNA. This means that the sequence of DNA in regulatory genes
affects phenotype and creates variation. Therefore, regulatory genes are subject to nat-
ural selection, just as protein coding genes are.
In humans and other eukaryotes, the expression of a given protein coding gene is
often affected by many regulatory sequences, which are sometimes located quite far
from the coding sequence that they regulate. The proteins that bind to these regula-
tory genes can interact in complex ways, so that multiple proteins bound to sequences
of DNA at widely separate sites may act to activate or repress a particular protein cod-
ing gene. However, the basic principle is usually the same: under some circumstances
the DNA sequence in the regulatory gene binds to a protein that in turn affects the
expression of a particular protein coding gene.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 49
Molecular Genetics 49
Complex, multicellular organisms are made of many different kinds of cells, each
with its own specific chemical machinery: cells in the liver and pancreas secrete diges-
tive enzymes, nerve cells carry electrical signals throughout the body, muscle cells
respond to these signals by performing mechanical work, and so on. Nonetheless, in
a single individual, all cells have exactly the same genetic composition. The cells dif-
fer in their function because different genes are activated in different cell types.
The development of the vertebrate nervous system provides an example of how
this works. At a certain point in the development of all vertebrate embryos, cells that
are going to give rise to the spinal cord differentiate to form the neural tube. Special
cells at one end of this structure secrete a molecule called “Sonic Hedgehog” (after the
video game character of the same name). Cells close to the source of this signaling
molecule experience a high concentration of Sonic Hedgehog; those more distant expe-
rience a lower concentration. The concentrations of Sonic Hedgehog affect the expres-
sion of genes in these future nerve cells. Low concentrations lead to the expression of
genes that destine the cells to become motor neurons, which control muscles; higher
concentrations lead to the expression of genes that cause cells to become the neurons
of the brain and spinal cord.
A single signal can trigger the complex series of events that transform a cell into
a liver cell or a nerve cell because the expression of one gene causes a different set of
genes to be expressed, and this in turn causes additional sets of genes to be expressed.
For example, a gene called Pax6 codes for a regulatory protein that is important in the
differentiation of the cells in the developing eye. Artificially activating this gene in one
cell in a fruit fly antenna leads to the synthesis of a regulatory protein that sets off a
cascade of gene expression, which eventually involves the expression of about 2,500
other genes and the development of an extra eye on the fly’s antenna.
Messenger RNA is not the only important form of RNA. Some RNA molecules bind
together with proteins to perform a variety of cellular functions. Ribosomes are an
example, as are spliceosomes, the organelles that splice the mRNA in eukaryotes after
the introns have been snipped out. It has long been known that some DNA codes for
these RNAs, but it was thought that the bulk of the 98% of the genome that lies out-
side of exons was just “junk DNA” with no real function. However, recent studies
have revealed that at least half of the genome is expressed as noncoding RNA
(ncRNA). While the function of much of the ncRNA is still mysterious, one form of
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 50
50 CHAPTER 2 • Genetics
ncRNA, short RNA segments called microRNAs (miRNAs), plays an important role
in regulating the translation of mRNA into protein. DNA sequences coding for
miRNA occur between coding sequences and in introns. These sequences are tran-
scribed, and the resulting miRNAs move outside the nucleus, where, after some pro-
cessing, they bind to complementary mRNA molecules and prevent them from being
translated into protein, providing an alternative mechanism for regulation of gene
expression. Much like regulatory sequences, some miRNAs play an important role in
regulating development and cell differentiation in complex organisms like humans.
Introns are not the only kind of DNA that is not involved in the synthesis of pro-
teins. Chromosomes also contain a lot of DNA that is composed of simple repeated
patterns. For example, in fruit flies there are long segments of DNA composed of ade-
nine and thymine in the following monotonous five-base pattern:
. . . ATAATATAATATAATATAATATAATATAATATAATATAAT . . .
In all eukaryotes, simple repeated sequences of DNA are found in particular sites on
particular chromosomes.
At this point, there is some danger of losing sight of the genes amid the
discussion of introns, exons, and repeat sequences. A brief reprise may be
helpful.
FURTHER READING
Chen, K., and N. Rajewsky. 2007. The evolution of gene regulation by transcription
factors and microRNAs. Nature Reviews: Genetics 8:93–103.
Maynard Smith, J. 1998. Evolutionary Genetics. 2nd ed. New York: Oxford Univer-
sity Press.
Olby, R. C. 1985. Origins of Mendelism. 2nd ed. Chicago: University of Chicago
Press.
Ridley, M. 1996. Evolution. 2nd ed. Cambridge, Mass.: Blackwell Science.
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 51
Key Terms 51
Snustad, D. P., and M. J. Simmons. 2006. Principles of Genetics. 4th ed. Hoboken,
N.J.: Wiley.
Watson, J. D., T. A. Baker, S. P. Bell, A. Gann, M. Levine, and R. Losick. 2008. Molec-
ular Biology of the Gene. 6th ed. San Francisco: Pearson/Benjamin Cummings.
Xing, Y., and C. Lee. 2006. Alternative splicing and RNA selection pressure—evolutionary
consequences for eukaryotic genomes. Nature Reviews: Genetics 7:499–509.
KEY TERMS
variants genes mitosis
crosses gametes diploid
F0 generation independent assortment homologous pairs
F1 generation chromosome meiosis
F2 generation nucleus haploid
02_Genetics [4p]:How Humans Evolved [5e] 10/10/08 11:08 AM Page 52
52 CHAPTER 2 • Genetics
Chapter 3
Population Genetics
Genes in Populations
How Random Mating and
Sexual Reproduction Change
Genotypic Frequencies
How Natural Selection
Changes Gene Frequencies
The Modern Synthesis
The Genetics of Continuous Variation
How Variation Is Maintained
Natural Selection and Behavior
Constraints on Adaptation
Correlated Characters
Disequilibrium
Genetic Drift
Local versus Optimal Adaptations
Other Constraints on Evolution
POPULATION GENETICS
Evolutionary change in a phenotype reflects change in the underlying genetic composition of the population.
W
hen we discussed Mendel’s experiments in Chapter 2, we briefly intro-
duced the distinction between genotype and phenotype: phenotypes are
the observable characteristics of organisms, and genotypes are the
underlying genetic compositions. There need not be a one-to-one correspondence
between genotype and phenotype. For example, there were only two seed-color phe-
notypes in Mendel’s pea populations (yellow and green), but there were three geno-
types (AA, Aa, and aa; Figure 3.1).
53
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 54
From what we now know about the genetic nature of inheritance, it is also clear
that evolutionary processes must entail changes in the genetic composition of popu-
lations. When evolution alters the morphology of a character like the finch’s beak,
there must be a corresponding change in the distribution of genes that control beak
development within the population. To understand how Mendelian genetics solves
each of Darwin’s difficulties, we need to look more closely at what happens to genes
in populations that are undergoing natural selection. This is the domain of population
genetics.
Genes in Populations
Peas were a use-
FIGURE 3.1
ful subject for Biologists describe the genetic composition of a population by specifying the
Mendel’s botanical experiments because frequency of alternative genotypes.
they have a number of dichotomous
traits. For example, pea seeds are either It’s easiest to see what happens to genes in populations if we consider a trait that is
yellow or green, but not an intermediate controlled by one gene operating at a single locus on a chromosome. Phenylketonuria
color.
(PKU), a potentially debilitating, genetically inherited disease in humans, is determined
by the substitution of one allele for another at a single locus. Individuals who are
homozygous for the PKU allele are missing a crucial enzyme in the biochemical path-
way that allows people to metabolize the amino acid phenylalanine. If the disease is
not treated, phenylalanine builds up in the bloodstream of children with PKU and
leads to severe mental retardation. Fortunately, treatment is possible: people with PKU
raised on a special low-phenylalanine diet can develop normally and lead normal lives
as adults.
How do evolutionary processes control the distribution of the PKU allele in a pop-
ulation? The first step in answering this question is to characterize the distribution of
the harmful allele. Population geneticists do this by specifying the genotypic frequency,
which is simply the fraction of the population that carries that genotype. Let’s label
the normal allele A and the deleterious PKU allele a. Suppose we perform a census of
a population of 10,000 individuals and determine the number of individuals with each
genotype. Geneticists can do this using established biochemical methods for deter-
mining an individual’s genotype for a specific genetic locus. Table 3.1 shows the num-
ber of individuals with each genotype and the frequencies of each genotype in this
hypothetical population. (In most real populations, the frequency of individuals
homozygous for the PKU allele is only about 1 in 10,000. We have used larger num-
bers here to make the calculations simpler.)
Genotypic frequencies must add up to 1.0 because every individual in the popu-
TABLE 3.1 The distribution of individuals with each of the three genotypes in a popula-
tion of 10,000.
Population Genetics 55
lation has to have a genotype. We keep track of the frequencies of genotypes, rather
than the numbers of individuals with each genotype, because the frequencies provide
a description of the genetic composition of populations that is independent of popu-
lation size. This makes it easy to compare populations of different sizes.
A variety of events in the lives of plants and animals act to change the frequency
of alternative genotypes in populations from generation to generation. Population
geneticists categorize these processes into a number of evolutionary mechanisms, or
“forces.” The most important mechanisms are sexual reproduction, natural selection,
mutation, and genetic drift. In the remainder of this section we will see how sexual
reproduction and natural selection alter the frequencies of genes and genotypes, and
later in the chapter we will return to consider the effects of mutation and genetic
drift.
The events that occur during sexual reproduction can lead to changes in
genotypic frequencies in a population.
First let’s consider the effects of the patterns of inheritance that Mendel observed.
Imagine that men and women do not choose their mates according to whether or not
they are afflicted with PKU, but mate randomly with respect to the individual’s geno-
type for PKU. It is important to study the effects of random mating because, for most
genetic loci, mating is random. Even though humans may choose their mates with care,
and might even avoid mates with particular genetic characteristics, such as PKU, they
cannot choose mates with a particular allele at each locus because there are about
30,000 genetic loci in humans. Random mating between individuals is equivalent to
the random union of gametes. In this sense, it’s not really different from oysters shed-
ding their eggs and sperm into the ocean, where chance dictates which gametes will
form zygotes.
p and the frequency of a as q. (Because there are only two alleles, p + q = 1.) If all indi-
viduals produce the same number of gametes, then we can calculate q as follows:
no. of a gametes
q5
total no. of gametes
Note that this is simply the definition of a frequency. We can calculate the values
of the numerator and denominator of this fraction from information we already know.
Because a gametes can be produced only by aa and Aa individuals, the total number
of a gametes is simply the sum of the number of Aa and aa parents multiplied by the
number of a gametes that each parent produces. The denominator is the number of
gametes per parent multiplied by the total number of parents. Hence,
1 2 1 2
no. of a no. of a
gametes per
aa parent
1 parents2
no. of aa
1 gametes per
Aa parent
1no.parents
of Aa
2
q5 (3.1)
(no. of gametes per parent)(total no. of parents)
To simplify this equation, let’s first examine the terms that involve numbers of
individuals. Remember that the population size is 10,000 individuals. This means that
the number of aa parents is equal to the frequency of aa parents multiplied by 10,000.
Similarly, the number of Aa parents is equal to the frequency of Aa parents multiplied
by 10,000. Now we examine the terms that involve numbers of gametes. Suppose each
parent produces two gametes. For aa individuals, both gametes contain the a allele,
so the number of a gametes per aa parent is 2. For Aa individuals, half of the gametes
will carry the a allele and half will carry the A allele, so here the number of a gametes
per Aa parent is 0.5 ✕ 2. Now we substitute all these values into Equation (3.1) to get
We can reduce this fraction by dividing the top and bottom by 2 ✕ 10,000, which yields
the following formula for the frequency of the a allele in the pool of gametes:
Notice that this form of the formula contains neither the population size nor the
average number of gametes per individual in the population. Under normal circum-
stances, the population size and the average number of gametes per individual do not
matter. This means that you can use this expression as a general formula for calculating
gene frequencies among the gametes produced by any population of individuals, as
long as the genetic locus of interest has only two alleles. It is important to keep in mind
that the formula results from applying Mendel’s laws and from counting the number
of a gametes produced.
By using Equation (3.2) and values from Table 3.1, for this particular population
we get q = 0.2 + (0.5 ✕ 0.4) = 0.4. Because p + q must sum to 1.0, p (the frequency of
the A gametes) must be 0.6. Notice that the frequency of each allele in the pool of
gametes is the same as the frequency of the same allele among parents.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 57
Population Genetics 57
The next step is to calculate the frequencies of all the genotypes among the
zygotes.
Now that we have calculated the frequency of the PKU allele among the pool of
gametes, we can determine the frequencies of the genotypes among the zygotes. If each
zygote is the product of the random union of two gametes, the process of zygote for-
mation can be schematically represented in an event tree (Figure 3.2) similar to the one
we used to represent Mendel’s crosses (for example, Figure 2.9). First we select a
gamete—say, an egg. The probability of selecting an egg carrying the a allele is 0.4
because this is the frequency of the a allele in the population. Now we randomly draw
a second gamete, the sperm. Again, the probability of getting a sperm carrying an a
allele is 0.4. The chance that these two randomly chosen gametes are both a is 0.4 ✕
0.4, or 0.16. Figure 3.2 also shows how to compute the probabilities of the other two
genotypes. Note that the sum of the three genotypes always equals 1.0, because all
individuals must have a genotype.
If we form a large number of zygotes by randomly drawing gametes from the
gamete pool, we will obtain the genotypic frequencies shown in Table 3.2. Compare
the frequencies of each genotype in the parental population (Table 3.1) with the fre-
quencies of each genotype in the F1 generation: the genotypic frequencies have
changed because the processes of independent segregation of alleles into gametes and
TABLE 3.2 The distribution of genotypes in the population of zygotes in the F1 genera-
tion.
random mating alter the distribution of alleles in zygotes. As a result, genotypic fre-
quencies between the F0 generation and the F1 generation are altered. [Note, however,
that the frequencies of the two alleles, a and A, have not changed; you can check this
out using Equation (3.2).]
If no other processes act to change the distribution of genotypes, the set of geno-
typic frequencies in Table 3.2 will remain unchanged in subsequent generations. That
is, if members of the F1 generation mate at random, the distribution of genotypes in
the F2 generation will be exactly the same as the distribution of genotypes in the F1 gen-
eration. The fact that genotypic frequencies remain constant was recognized inde-
pendently by British mathematician Godfrey Harold Hardy and German physician
Wilhelm Weinberg in 1908, and these constant frequencies are now called the Hardy–
Weinberg equilibrium. As we will see later in this chapter, the realization that it’s not
just sexual reproduction alone that alters phenotypic and genotypic frequencies was
the key to understanding how variation is maintained.
In general, the Hardy–Weinberg proportions for a genetic locus with two alleles
are
freq(aa) = q2
freq(Aa) = 2pq
freq(AA) = p2 (3.3)
where q is the frequency of allele a, and p is the frequency of allele A [Equation (3.3)].
Using a Punnett square, Figure 3.3 shows how to calculate these frequencies.
If no other processes act to change genotypic frequencies, the Hardy–Weinberg
equilibrium frequencies will be reached after only one generation and will remain
unchanged thereafter. Moreover, if the Hardy–Weinberg proportions are altered by
Female gametes
freq(A) = p freq(a) = q= 1 – p
p 1–p
A A A A A A a a a a
A
A
This Punnett square shows a second way
FIGURE 3.3 A A + A = AA A + a = aA
to calculate the frequency of each type
of zygote when there is random mating. The horizontal axis rep- freq(A) = p A
resents the proportion of A and a eggs, and is thus divided into freq(AA) = p2
Male gametes
A freq(aA) = p (1 – p)
fractions p and q, where q = 1 – p. The vertical axis is divided
A
according to the proportion of sperm carrying each allele, and
again it is divided into fractions p and q. The areas of the rec- A
tangles formed by the intersection of the vertical and horizontal a
dividing lines give the proportion of zygotes that result from a + A = Aa a + a = aa
each of the four possible fertilization events. The area of the a
freq(a)=q = 1 – p
square containing aa zygotes is q2, the area of the square con- a freq(aA) = p(1 – p) freq(aa) = (1 – p)2
taining AA zygotes is p2, and the total area of the two rectan-
a
gles containing Aa zygotes is 2pq.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 59
Box 3.1
Genotypic Frequencies
after Two Generations
of Random Mating
From Equation (3.2) given in the text, we know that the fre- Because there are only two alleles, the frequency of A is 0.6.
quency of the gametes produced by adults in the second gen- As we asserted in the text, the frequency of the two alleles
eration carrying the a allele will be remains unchanged, and the frequencies of the gametes
produced by random mating among members of this gen-
q = freq(aa) + 0.5 ✕ freq(Aa) eration are the same as in the previous generation. The cal-
= 0.16 + 0.5 ✕ 0.48 culations for these frequencies are the same as those shown
= 0.16 + 0.24 in Table 3.2.
= 0.4
We have already calculated the frequency of each of the genotypes among zygotes
just after conception (see Table 3.2), but if PKU is a lethal disease without treatment,
not all of these individuals will survive. Now we need to calculate the frequency of a
gametes after selection, which we can do by expanding Equation (3.4) as follows:
1 21 2 1 21 2
no. of a no. of aa no. of a no. of Aa
gametes per parents gametes per parents
1
aa parent after after Aa parent after after
selection selection selection selection
q9 5 (3.5)
(no. of gametes per parent)(total no. of parents after selection)
Because none of the aa individuals will survive, the first term in the numerator will be
equal to zero. If we assume that all of the AA and Aa parents survive, then the num-
ber of parents after selection is 10,000 ✕ [freq(AA) + freq(Aa)] and thus Equation (3.5)
can be simplified to
This calculation shows that the frequency of the PKU allele in adults is 0.2857, a big
decrease from 0.4.
Several important lessons can be drawn from this example:
• Selection cannot produce change unless there is variation in the population. If all
the individuals were homozygous for the normal allele, gene frequencies would
not change from one generation to the next.
• Selection does not operate directly on genes and does not change gene frequencies
directly. Instead, natural selection changes the frequency of different phenotypes.
In this case, individuals with PKU cannot survive without treatment. Selection
decreases the frequency of the PKU allele because it is more likely to be associated
with the lethal phenotype.
• The strength and direction of selection depend on the environment. In an environ-
ment with medical care, the strength of selection against the PKU allele is negligible.
It is also important to see that although this example shows how selection can
change gene frequencies, it does not yet show how selection can lead to the evolution
of new adaptations. Here, all phenotypes were present at the outset; all selection did
was change their relative frequency.
on the shape of the pea seeds and the height of the pea plants. Genetics seemed to
prove that inheritance was fundamentally discontinuous, and early twentieth-century
geneticists like Hugo de Vries and William Bateson argued that this fact could not be
reconciled with Darwin’s idea that adaptation occurs through the accumulation of
small variations. If one genotype produces short plants and the other two genotypes
produce tall plants, then there will be no intermediate types, and the size of pea plants
cannot change in small steps. In a population of short plants, tall ones must be cre-
ated all at once by mutation, not gradually lengthened over time by selection. Most
biologists of the time found these arguments convincing, and consequently Darwin-
ism was in decline during the early part of the twentieth century.
In the early 1930s, the British biologists Ronald A. Fisher and J. B. S. Haldane and
the American biologist Sewall Wright showed how Mendelian genetics could be used
to explain continuous variation (Figure 3.4). We will see how their insights led to the
resolution of the two main objections to Darwin’s theory: (1) the absence of a theory
of inheritance, and (2) the problem of accounting for how variation is maintained in
populations. When the theory of Wright, Fisher, and Haldane was combined with Dar-
win’s theory of natural selection, and with modern field studies by biologists such as
Theodosius Dobzhansky, Ernst Mayr, and George Gaylord Simpson, a powerful
explanation of organic evolution emerged. This body of theory, and the supporting
empirical evidence, is now called the modern synthesis.
Continuously varying characters are affected by genes at many loci, each locus
having only a small effect on phenotype.
To see how the theory of Wright, Fisher, and Haldane works, let’s start with an
unrealistic, but instructive, case. Suppose that there is a measurable, continuously vary-
ing character such as beak depth, and suppose that two alleles, + and –, operating at
a single genetic locus control the character. We’ll assume that the gene at this locus
Shown here are the three architects of the modern synthesis, which showed how
FIGURE 3.4
Mendelian genetics could be used to account for continuous variation: (a) Ronald
A. Fisher, (b) J. B. S. Haldane, and (c) Sewall Wright.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 62
influences the production of a hormone that stimulates beak growth, and that each
allele leads to production of a different amount of the growth hormone. Let’s say that
each “dose” of the + allele increases the beak depth, while a – dose decreases it. Thus,
+ + individuals have the deepest beaks, – – individuals have the shallowest beaks, and
+ – individuals have intermediate beaks. In addition, suppose the frequency of the +
allele in the population is 0.5. Now we use the Hardy–Weinberg rule [Equation (3.3)]
to calculate the frequencies of different beak depths in the population. A quarter of
the population will have deep beaks (+ +), half will have intermediate beaks (+ –),
and the remaining quarter will have shallow beaks (– –; Figure 3.5).
This does not look like the smooth, bell-shaped distribution of beak depths that
the Grants observed on Daphne Major (see Chapter 1). If beak depth were controlled
by a single locus, natural selection could not increase beak depth in small increments.
But it turns out that the beak morphology is actually affected by several genes. Arhat
Abkazov, a Harvard molecular biologist, and the Grants have recently identified sev-
eral genes that affect beak morphology in Darwin’s finches. The level of expression of
one of these genes, Bmp4, affects beak depth, and the level of expression of two other
genes, Bmp2 and Bmp7, influences overall beak size. Scientists do not yet know
whether the levels of expression in these genes are due to mutations in associated reg-
ulatory sequences or mutations that affect other genes that regulate their expression.
But imagine what would happen if other genes at a second locus on a different chro-
mosome also affected beak depth, perhaps because they controlled the synthesis of the
receptors for the growth hormone. As before, we assume there is a + allele that leads
to larger beaks and a – allele that leads to smaller beaks. Using the Hardy–Weinberg
proportions and assuming the independent segregation of chromosomes, we can show
that there are now more types of genotypes, and the distribution of phenotypes begins
to look somewhat smoother (Figure 3.6). Now imagine that a third locus controls the
calcium supply to the growing beak, and that once again there is a + allele leading to
larger beaks and a – allele leading to smaller beaks. As Figure 3.7 shows, the distri-
bution of beak depths is now even more like the bell-shaped distribution seen in
nature. However, small gaps still exist in this distribution of beaks. If genes were the
only influence on beak depth, we might expect to see a more broken distribution than
the Grants actually observed in the Galápagos.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 63
Darwin’s view of natural selection is easily incorporated into the genetic view
that evolution typically results from changes in gene frequencies.
Darwin knew nothing about genetics, and his theory of adaptation by natural selec-
tion was framed in purely phenotypic terms: there is a struggle for existence, there is
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:21 AM Page 64
phenotypic variation that affects survival and reproduction, and this phenotypic vari-
ation is heritable. In Chapter 1, we saw how this theory could explain the adaptive
changes in beak depth in a population of Darwin’s finches on the Galápagos Islands. As
we saw earlier in this chapter, however, population geneticists take the seemingly dif-
ferent view that evolution means changes in allelic frequencies by natural selection. But
these two views of evolution are easily reconciled. Suppose that Figure 3.8 gave the dis-
tribution of beak depths before the drought on Daphne Major. Remember that indi-
viduals with deep beaks were more likely to survive the drought and reproduce than
individuals with smaller beaks were. Figure 3.8 illustrates that individuals with deep
beaks are more likely to have + alleles at the three loci assumed to affect beak depth.
Thus, at each locus, + + individuals have deeper beaks on average than + – individuals,
which in turn have deeper beaks on average than – – individuals. Because individuals
with deeper beaks had higher fitness, natural selection would favor the + alleles at each
of the three loci affecting beak depth, and thus the + alleles would increase in frequency.
F0 generation 3 3
Gametes
The blending model of inheritance assumes that
FIGURE 3.9
the hereditary material is changed by mating.
F1 generation 3 3 When red and white parents are crossed to produce a pink offspring,
the blending model posits that the hereditary material has mixed, so
that when two pink individuals mate, they produce only pink offspring.
Gametes
According to Mendelian genetics, however, the effects of genes are
blended in their expression to produce a pink phenotype, but the genes
F2 generation
themselves remain unchanged. Thus, when two pink parents mate, they
can produce white, pink, or red offspring.
one generation and the distribution of phenotypes does not change. Furthermore, we
know that sexual reproduction produces no blending in the genes, even though off-
spring may appear to be intermediate between their parents. This is because genetic
transmission involves faithful copying of the genes themselves and their reassembly in
different combinations in zygotes. The only blending that occurs takes place at the level
of the expression of genes in phenotypes. The genes themselves remain distinct phys-
ical entities (Figure 3.9).
These facts do not completely solve the problem of the maintenance of variation,
because selection tends to deplete variation. When selection favors birds with deeper
beaks, we might expect – alleles to be replaced at all three loci affecting the trait, leav-
ing a population in which every individual has the genotype + + + + + +. There would
still be phenotypic variation due to environmental effects, but without genetic varia-
tion there can be no further adaptation.
Genes are copied with amazing fidelity, and their messages are protected from ran-
dom degradation by a number of molecular repair mechanisms. Every once in a while,
however, a mistake in copying is made and goes unrepaired, and a new allele is intro-
duced into the population. In Chapter 2, we learned that genes are pieces of DNA. Cer-
tain forms of ionizing radiation (such as X-rays) and certain kinds of chemicals
damage the DNA and alter the message that it carries. These changes are called muta-
tions, and they add variation to a population by continuously introducing new alleles,
some of which may produce novel phenotypic effects that selection can assemble into
adaptations. Although rates of mutation are very low—ranging from 1 in 100,000
to 1 in 10 million per locus per gamete in each generation—this process plays an
important role in generating variation.
Low mutation rates can maintain variation because a lot of variation is protected
from selection.
For characters that are affected by genes at many different loci, low rates of muta-
tion can maintain variation in populations. This is possible because many different
genotypes generate intermediate phenotypes that are favored by stabilizing selection.
If individuals with a variety of genotypes are equally likely to survive and reproduce,
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 66
Hidden variation explains why selection can move populations far beyond their
initial range of variation.
Remember from Chapter 1 Fleeming Jenkin’s argument that Darwin’s theory
could not explain cumulative evolutionary change because it provided no account of
how a population could evolve beyond the initial range of variation present. Selection
would cull away all of the small-beaked finches, he would have argued, but it could
never make the average beak bigger than the biggest beak initially present. If this argu-
ment were correct, then selection could never lead to cumulative, long-term change.
Jenkin’s argument was wrong because it failed to take hidden variation into
account. Hidden variation is always present in continuously varying traits. Let’s sup-
pose that environmental conditions favor larger beaks. When beak depth is affected
by genes at many loci, the birds in a population with the deepest beaks do not carry
Segregation and
recombination
Mutation
1
Expressed Hidden
variation variation
18
16
14
Percentage of oil
12
10
all + alleles. They carry a lot of + alleles and some – alleles. When the finches with the
shallowest beaks die, alleles leading to small beaks are removed from the breeding pop-
ulation. As a result, the frequency of + alleles at every locus increases. But because even
the deepest-beaked individuals had some – alleles, a huge amount of variation
remains. This variation is shuffled through the process of sexual reproduction. Because
+ alleles become more common, more of these alleles are likely to be combined in the
genotype of a single individual. The greater the proportion of + alleles in an individ-
ual, the larger the beak will be. Thus, the biggest beak will be larger than the biggest
beak in the previous generation. In the next generation, the same thing happens again:
the deepest-beaked individuals still carry – alleles, but fewer than before. As a result,
the biggest beaks can be bigger than any beaks in the previous generation.
This process can go on for many generations. An experiment on oil content in
corn begun at the University of Illinois Agricultural Experiment Station in 1896 pro-
vides a good example. Researchers selected for high and low oil content in corn, and,
as Figure 3.11 shows, each generation showed significant change. In the initial popu-
lation of 163 ears of corn, oil content ranged from 4% to 6%. After nearly 80 gener-
ations of selection, both the high and low values for oil content far exceeded the initial
range of variation. Researchers were then even able to reverse the direction of adap- This Pekingese
FIGURE 3.12
tation by taking plants from the high-oil line and selecting for low oil content. is smaller than
An even more dramatic example is the Pekingese, whose very existence is a testa- the smallest wolf, although wolves are
the ancestors of all domestic breeds of
ment to the power of hidden variation (Figure 3.12). In a few thousand generations,
dogs. Some of Darwin’s critics argued
a mere instant on the evolutionary timescale, wolves were transformed into Pekingese
that natural selection could not produce
as well as into bulldogs, Great Danes, dachshunds, Chihuahuas, Irish wolfhounds, and
this kind of transformation because it
toy poodles. This is far too little time for a large number of new mutations to accu- could not lead to changes that fell out-
mulate. Instead, most of the genes necessary for the development of Irish wolfhounds side the range of variation found in the
and Pekingese must have been present in the ancestral population of wolves from which original population. However, genes for
dogs were selected, their effects hidden by other genes. Dog breeders, acting as agents very small size are present in wolves, but
of selection, chose the genes leading to small size, long hair, and funny little faces, and their effects are hidden by other genes.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 68
So far we have considered the evolution of morphological characters, like beak depth
and eye morphology, that do not change once an individual has reached adulthood.
In much of this book, we will be interested in the evolution of the behavior of humans
and other primates. Behavior is different from morphology in an important way: it is
flexible, and individuals adjust their behavior in response to their circumstances. Some
people think natural selection cannot account for flexible responses to environmen-
tal contingencies because natural selection acts only on phenotypic variation that
results from genetic differences. Although this view is very common (particularly
among social scientists), it is incorrect. To see why, let’s consider an elegant empirical
study that illustrates exactly how natural selection can shape flexible behavioral
responses.
The soapberry bug (Jadera haematoloma), a seed-eating insect found in the
southeastern United States, has been studied by biologist Scott Carroll of the Univer-
sity of New Mexico. (It’s OK to call them “bugs” because they are members of the
insect order Hemiptera, the true bugs.) Adult soapberry bugs are bright red and black,
and about 1 to 1.5 cm (0.5 in.) long (Figure 3.13). They gather in huge groups near
the plants they eat. During mating, males mount females and copulate. The transfer
of sperm typically takes about 10 minutes. However, males remain in the copulatory
position, sometimes for hours, securely anchored to females by large genital hooks.
This behavior is called mate guarding. Biologists believe that the function of mate
guarding is to prevent other males from copulating with the female before she lays her
eggs. When a female mates with several males, they share in the paternity of the eggs
that she lays. By guarding his mate and preventing her from mating with other males,
a male can increase his reproductive success. However, mate guarding also has a cost:
The male soap- a male cannot find and copulate with other females while guarding a mate. The rela-
FIGURE 3.13
berry bug on tive magnitude of the costs and benefits of mate guarding depends on the sex ratio (the
the left is guarding the (larger) female relative numbers of males and females). When the ratio of males to females is high,
below him. Biologist Scott Carroll
males have little chance of finding another female, and guarding is the best strategy.
painted numbers on the bugs in order to
When females are more common than males, the chance of finding an unguarded
identify individuals. (Photograph cour-
female increases, so males may benefit more from looking for additional mates than
tesy of Scott Carroll.)
from guarding.
In populations of soapberry bugs in western Oklahoma, the sex ratio is quite vari-
able. In some places, there are equal numbers of males and females; in others, there
are twice as many males as females. Males guard their mates more often where females
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 69
0.9
0.8
are rare than where they are common (Figure 3.14). Two possible mechanisms might
produce this pattern: (1) males in populations with high sex ratios might differ genet-
ically from males in populations with low sex ratios, or (2) males might adjust their
behavior in response to the local sex ratio. To distinguish between these two possi-
bilities, Carroll brought soapberry bugs into the laboratory and created populations
with different sex ratios. Then he watched males mate in each of these populations.
If the mate-guarding trait were canalized (that is, showed the same phenotype in a wide
range of environments), then a male would behave the same way in each population;
if the trait were plastic, then a male would adjust his behavior in relation to the local
sex ratio. Carroll’s data confirmed that soapberry bug males in Oklahoma have a plas-
tic behavioral strategy that causes them to modify their mating behavior in response
to current social conditions.
Evidence suggests that the soapberry bug’s plasticity has evolved in response to
the variability in conditions in Oklahoma.
Most soapberry bugs live south of Oklahoma, in warmer, more stable habitats like
the Florida Keys, and the sex ratio in these areas is always close to even. (Hence, females
are relatively rare.) Carroll subjected male soapberry bugs from the Florida Keys to the
same experimental protocol as the bugs from Oklahoma. As Figure 3.15 shows, the
Florida males did not change their behavior in response to changes in sex ratio. They
guarded their mates about 90% of the time, regardless of the abundance of females. In
a stable environment like the Florida Keys, sex ratios do not vary much from time to
time, and the ability to adjust mate-guarding behavior provides no advantage. Behav-
ioral flexibility is costly in a number of ways. For example, flexible males must spend
time and energy assessing the sex ratio before they mate, males sometimes make mis-
takes about the local sex ratio and behave inappropriately, and flexibility probably
requires a more complex nervous system. Thus, a simple fixed behavioral rule is likely
to be best in stable environments. In the variable climate of Oklahoma, however, the
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 70
1.0
Oklahoma
0.8
0.7
1:2 1:1 2:1
Sex ratio (M:F)
1.0
Florida
0.8
The average probability of mate guarding varies
FIGURE 3.15
between populations of soapberry bugs in Oklahoma
and Florida. In Oklahoma, males are more likely to guard when females are 0.7
relatively rare; in Florida, males do not vary their mate-guarding behavior in 1:2 1:1 2:1
relation to the local sex ratio. Sex ratio (M:F)
How did behavioral flexibility evolve in the Oklahoma bugs? It evolved like any
other adaptation: by the selective retention of beneficial genetic variants. For any char-
acter to evolve, (1) the character must vary, (2) the variation must affect reproductive
success, and (3) the variation must be heritable. Mate guarding in soapberry bugs sat-
isfies each of these conditions: First, there is variation, and this variation affects fitness.
By bringing individual bugs into the laboratory and observing their behavior at dif-
ferent sex ratios, Carroll showed that individual males in Oklahoma have different
behavioral strategies. Figure 3.16 plots the probability of mate guarding for four rep-
resentative individuals numbered 1 to 4. Males 1 and 4 both have fixed behavioral
strategies. Male 1 guards about 90% of the time, and male 4 guards about 80% of the
time. Males 2 and 3 have variable behavioral strategies. Male 2 is very sensitive to
changes in sex ratio; male 3 is less sensitive. Notice that both the amount of mate
guarding and the amount of flexibility vary among the bugs in Oklahoma.
Second, it seems likely that this variation would affect male reproductive success.
In Oklahoma, males would experience a range of sex ratios, so it seems likely that
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 71
Constraints on Adaptation 71
1.0
Male 1
Probability of mate guarding
0.9
Male 2
Male 3
Male 4
0.8 There is genetic variation in the behavioral
FIGURE 3.16
rules of individual soapberry bugs, illustrated
here by four representative individuals. There is variation in both the
level of mate guarding (for example, male 1 spends more time guard-
ing mates than male 4 does for all sex ratios) and in the amount of
plasticity (for example, the behavior of males 1 and 4 does not
change, the behavior of male 3 changes a bit, and the behavior of
0.7 male 2 changes a lot). If this variation is heritable, the rule that works
1:2 1:1 2:1 best, averaged over all of the environments of the population, will
Sex ratio (M:F) tend to increase.
males with flexible strategies, like male 2, would tend to have the most offspring. In
Florida, males with inflexible strategies, like male 1, would have the most offspring.
Third, the character is heritable. By controlling matings in the laboratory, Carroll
was able to show that males tended to have the same strategies as their fathers. Bugs
like male 1 had sons with fixed strategies, and bugs like male 2 had sons with flexible
strategies. Thus, the Oklahoma bugs would come to have a variable strategy, while the
Florida bugs would come to have a fixed strategy.
Behavior in the soapberry bug is relatively simple. Mate guarding depends on the
sex ratio in the local population. The behavior of humans and other primates is much
more complex, but the principles that govern the evolution of complex forms of behav-
ior are the same as the principles that govern the evolution of simpler forms of behav-
ior. That is, individuals must differ in the ways they respond to the environment, these
differences must affect their ability to survive and reproduce, and at least some of these
differences must be heritable. Then individual responses to environmental circum-
stances will evolve in much the same way that finch beaks and male soapberry bugs’
mate-guarding behavior evolve.
CONSTRAINTS ON ADAPTATION
Natural selection plays a central role in our understanding of evolution because it is
the only mechanism that can explain adaptation. However, evolution does not always
lead to the best possible phenotype. In this section, we consider five reasons why this
is the case.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 72
Correlated Characters
When individuals that have particular variants of one character also tend to have
particular variants of a second character, the two characters are said to be
correlated.
So far, we have considered the action of natural selection on only one character at a
time. This approach is misleading if natural selection acts on more than one charac-
ter simultaneously and the characters are nonrandomly associated, or correlated. It’s
easiest to grasp the meaning and importance of correlated characters in the context of
a now familiar example: Darwin’s finches. When the Grants and their colleagues cap-
tured medium ground finches on Daphne Major, they measured beak depth, beak
width, and a number of other morphological characters. Beak depth is measured as the
top-to-bottom dimension of the beak; beak width is the side-to-side dimension. As is
common for morphological characters like these, the Grants found that beak depth
and beak width are positively correlated: birds with deep beaks also have wide beaks
(Figure 3.17). Each point in Figure 3.17 represents one individual. Beak width is plot-
ted on the vertical axis, and beak depth is plotted along the horizontal axis. If the cloud
of points were round or the points were randomly scattered in the graph, it would
mean these characters were uncorrelated. Then information about an individual’s beak
depth would tell us nothing about its beak width. However, the cloud of points forms
an ellipse with the long axis oriented from the lower left to the upper right, so we know
that the two characters are positively correlated: deep beaks also tend to be wide. If
birds with deep beaks tended to have narrow beaks, and birds with shallow beaks
tended to have wide beaks, the two characters would be negatively correlated, and the
long axis of the cloud of points would run from the upper left to the lower right.
Correlated characters occur because some genes affect more than one
character.
Genes that affect more than one character are said to have pleiotropic effects, and
probably most genes fall into this category. Often genes that are expressed early in
Beak width
Constraints on Adaptation 73
development and affect overall size also influence a number of other discrete mor-
phological traits. This means that individuals carrying mutations that lead to increased
expression in the genes that influence overall beak size, Bmp2 and Bmp7, will have
deeper and wider beaks than individuals who don’t carry these mutations. The PKU
locus provides another good example of pleiotropy. Untreated PKU homozygotes are
phenotypically different from heterozygotes or normal homozygotes in several ways.
For example, they have lower IQs and different hair color from individuals with other
genotypes. Thus, the substitution of another PKU allele for the normal allele in a het-
erozygote would affect a wide variety of phenotypic characters.
When two characters are correlated, selection that changes the mean value of
one character in the population also changes the mean value of the correlated
character.
Returning to the finches, suppose that there is selection for individuals with deep
beaks, and that beak width has no effect on survival (Figure 3.18). As we would
expect, the mean value of beak depth increases. Notice, however, that the mean value
of beak width also increases, even though beak width has no effect on the probabil-
ity that an individual will survive. Selection on beak depth affects the mean value of
beak width because the two traits are correlated. This effect is called the correlated
response to selection. It results from the fact that selection increases the frequency of
genes that increase both beak depth and beak width.
To understand how selection on one character can cause other characters to change
in a maladaptive (less fit) direction, let’s continue with our example. It turns out that
beak width did affect survival during the Galápagos drought. By holding beak depth
After selection
Before selection
constant, the Grants showed that individuals with thinner beaks were more likely
to survive during the drought, probably because birds with thinner beaks were able to
generate more pressure on the tough seeds that predominated during the drought. If
selection were acting only on beak width, then we would expect the mean beak width
in the population to decrease. However, the correlated response to selection on beak
depth also acts to increase mean beak width. If the correlated response to selection on
beak depth were stronger than the effect of selection directly on beak width, mean beak
width would increase, even though selection favors thinner beaks. This is exactly what
happened on Daphne Major (Figure 3.19).
Disequilibrium
In Chapter 1, we saw how natural selection could gradually increase beak depth, gen-
eration by generation, until an equilibrium was reached—a point where stabilizing
selection maintained the average beak depth in the population at the optimum size. This
example illustrates the principle that selection keeps changing a population until an
adaptive equilibrium is reached. It is easy to forget that the populations being observed
have not necessarily reached equilibrium. If the environment has changed recently, there
is every reason to suspect that the morphology or behavior of residents is not adaptive
under current conditions. How long does it take for populations to adapt to environ-
mental change? This depends on how quickly selection acts. As we have seen, enormous
changes can be created by artificial selection in a few dozen generations.
Disequilibrium is particularly important for some human characters because there
have been big changes in the lives of humans during the last 10,000 years. It seems
likely that many aspects of the human phenotype have not had time to catch up with
recent changes in our subsistence strategies and living conditions. Our diet provides
Threshold Wide-beaked
width birds die
Mean beak
depth and width:
Beak width
After selection
Before selection
Constraints on Adaptation 75
So far, we have assumed that evolving populations are always very large. When pop-
ulations are small, however, random effects caused by sampling variation can be
important. To see what this means, consider a statistical analogy. Suppose that we have
a huge urn like the one in Figure 3.21. The urn contains 10,000 balls—half of them
black and half of them red. Suppose we also have a collection of small urns, each of
which holds 10 balls. We draw 10 balls at random from the big urn to put in each
small urn. Not all of the little urns will have five red balls and five black balls. Some
will have four red balls, some three red balls, and a few may even have no red balls.
The fact that the distribution of black and red balls among the small urns varies is
called sampling variation.
The same thing happens during genetic transmission in small populations (Figure
3.22). Suppose there is an organism that has only one pair of chromosomes with two
possible alleles, A and a, at a particular locus and that selection does not act on this
trait. In addition, a population of five individuals of this species is newly isolated from
the rest. In this small population (generation 1), each allele has a frequency of 0.5, so
five chromosomes carry A and five carry a. These five individuals mate at random and
produce five surviving offspring (generation 2). To keep things simple, assume that the
gamete pool is large, so there is no sampling variation. This means that half of the
gametes produced by generation 1 will carry A and half will carry a. However, only
10 gametes will be drawn from this gamete pool to form the next generation of five
individuals. These gametes will be sampled in the same way the balls were sampled
from the urn (see Figure 3.21). The most likely result is that there are five A alleles and
five a alleles in generation 2, as there were in generation 1. However, just as there is
some chance of drawing three, four, six, seven, or even zero black balls from the urn,
there is some chance that generation 2 will not carry equal numbers of A and a alleles.
Suppose the five individuals in generation 2 have six A alleles and four a alleles. Since
the frequency of the A allele is now 0.6, when these individuals mate, the frequency
of the A allele in the pool of gametes will be 0.6 as well. This means that the frequency
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 76
Draw 10 balls
at random
of the allele in the population has changed by chance alone. In the third generation the
gene frequencies change again, this time in the opposite direction. This phenomenon
is called genetic drift. In small populations, genetic drift causes random fluctuations
in genetic frequencies.
The way alleles are sampled in real populations depends on the biology of the
evolving species and the nature of the selective forces that the species faces. For
instance, population size in some bird species might be limited by nesting sites, and
success in obtaining nesting sites might be related to body size: perhaps the larger the
bird, the more chance there would be that it would have success in battling others for
a prime nesting spot. Although body size is influenced partly by genes, and selection
would favor genes for large body size, many genes don’t affect body size, and they may
be sampled randomly in this particular population.
Genetic drift causes more rapid change in small populations than in large ones
because sampling variation is more pronounced in small samples. To see why this is
true, consider a simple example. Suppose a company is trying to predict whether a new
product—say, low-fat peanut butter—will succeed in the marketplace. The company
commissions two different survey firms to find out how many peanut-butter fans
would switch to the new product. One firm conducts interviews with five people, four
of whom say that they would switch immediately to the low-fat variety. The other
company polls 1,000 people, and 50% of them say they would prefer low-fat peanut
butter. Which survey should the company trust? Clearly, the second survey is more
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 77
Constraints on Adaptation 77
a a Generation 1
A a AA A a A a
5 black : 5 white
A a AA A a A a A a Generation 2
6 black : 4 white
Sampling varia-
FIGURE 3.22
tion leads to
changes in gene frequency in small pop-
ulations. Suppose that a population pro-
Very large gamete pool duces half A gametes (black) and half
a gametes (white), and the next gener-
60% black : 40% white ation consists of just five individuals. It is
not unlikely that, by chance, these indi-
viduals will carry six A gametes and four
a gametes. This new population will pro-
duce 60% A gametes. Thus, sampling
variation can change gene frequencies.
Generation 3 Notice here that gene frequencies
A a a a a a A a A a
3 black : 7 white change again in generation 3, this time
in the reverse direction, to 30% A.
credible than the first one. It is easy to imagine that by chance we might find four fans
of low-fat peanut butter in a sample of five people, even if the actual frequency of such
preferences in the population were only 50%. In contrast, in a sample of 1,000 peo-
ple, we would be very unlikely to find such a large discrepancy between our sample
and the population.
The same principle applies to genetic drift. In a population of five individuals (with
10 chromosomes) in which two alleles are equally common (frequency = 0.5), there
is a good chance that in the next generation the frequency of one of the alleles will be
greater than or equal to 0.8. But in a population of 1,000 individuals, there is virtu-
ally no chance that such a large deviation from the initial frequency will occur.
Genetic drift leads to unpredictable evolution because the changes in gene fre-
quency caused by sampling variation are random. As a result, drift causes isolated pop-
ulations to become genetically different from one another over time. Results from
a computer simulation illustrate how this works for a single genetic locus with two
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 78
This computer
FIGURE 3.23 1.0
simulation
demonstrates that genetic drift causes
isolated populations to become geneti-
cally different from one another. Four 0.8
populations are initially formed by 20
individuals being drawn from a single
Frequency of A
large population. In the original popula- 0.6
tion, one gene has two alleles that each
have a frequency of 0.5. The graph
tracks the frequency of one of these
0.4
alleles in each of these populations over
time. Population 1 is in red (solid cir-
cles), population 2 in blue (open trian-
gles), population 3 in green (solid 0.2
triangles), and population 4 in purple
(open circles). The dotted line shows
the global frequency of the A allele. In 0.0
each population, the frequency of the 0 5 10 15 20 25 30
allele fluctuates randomly under the Generation
influence of genetic drift. Notice that
populations 2 and 4 eventually reach
fixation, meaning that one of the two
alleles is lost. alleles (Figure 3.23). Initially, there is a large population in which each allele has a
frequency of 0.5. Then four separate populations of 20 individuals (each individual
carrying two sets of chromosomes) are created. These four populations are maintained
at a constant size and remain isolated from each other during all subsequent genera-
tions. During the first generation, the frequency of A in the gamete pool is 0.5 in each
population. However, the unpredictable effects of sampling variation alter the fre-
quencies of A among adults in each population. From the first generation, we sample
40 gametes from each of the four populations. In population 1 (marked as red solid
circles), the frequency of A increases dramatically; in populations 2 (blue open trian-
gles) and 3 (green solid triangles), it increases a little; and in population 4 (purple open
circles), it decreases. In each case, the change is created by chance alone.
During the next generation, the same process is repeated, except that the gamete
pools of each of the four populations now differ. Once again, 40 gametes are sampled
from each population. This time the frequency of A in populations 2, 3, and 4
increases, while the frequency of A in population 1 decreases substantially. As these
random and unpredictable changes continue, the four populations become more and
more different from one another. Eventually, one of the alleles is lost entirely in two
of our populations (A in population 2 and a in population 4), making all individuals
in these populations identical at the locus in question. Such populations are said to
have reached fixation. In general, the smaller the population, the sooner it reaches fix-
ation. Theoretically, if genetic drift goes on long enough, all populations eventually
reach fixation. When populations are large, however, this may take such a long time
that the species becomes extinct before fixation occurs. Populations remain at fixation
until mutation introduces a new allele.
As you might expect, the rate at which populations become genetically different
is strongly affected by their size: small populations differentiate rapidly; larger pop-
ulations differentiate more slowly. We will see in Chapter 13 that the relationship
between population size and the rate of genetic drift allows us to make interesting
and important inferences about the size of human populations tens of thousands of
years ago.
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 79
Constraints on Adaptation 79
Natural selection acts to increase the adaptedness of populations, but it does not nec-
essarily lead to the best possible phenotypes. The reason is that natural selection is
myopic: it favors small improvements to the existing phenotype, but it does not take
into account the long-run consequences of these alterations. Selection is like a moun-
taineer who tries to climb a peak cloaked in dense cloud; she cannot see the surround-
ing country, but she figures she will reach her goal if she keeps climbing uphill. The
mountain climber will eventually reach the top of something if she continues climb-
ing, but if the topography is the least bit complicated, she won’t necessarily reach the
summit. Instead, when the fog clears, she is likely to find herself on the top of a small,
subsidiary peak. In the same way, natural selection keeps changing the population until
no more small improvements are possible, but there is no reason to believe that the end
product is the optimal phenotype. The phenotype arrived at in such cases is called a
“local adaptation”; it is analogous to the subsidiary peak reached by the mountaineer.
As an example of this effect, let’s consider the evolution of eyes (Figure 3.24).
Humans, other vertebrates, and some invertebrates such as octopi have camera-type
eyes. In this type of eye, there is a single opening in front of a lens, which projects an
image on photoreceptive tissue. Insects, by contrast, have compound eyes, in which
many very small separate photoreceptors build up an image composed of a grid of
dots, something like a television image. Compound eyes are inferior to camera-type
eyes in most ways. For example, they have lower resolution and less light-gathering
power than similarly sized camera-type eyes have. If this is so, why haven’t insects
evolved camera-type eyes?
The most likely answer is that once a species has evolved complex, compound
eyes, selection cannot favor intermediate types, even though this might eventually
allow superior camera-type eyes to evolve. Consider the early lineages in which eyes
were first evolving. There may have come a time when selection favored greater light
sensitivity. In the vertebrates and mollusks, greater light sensitivity was achieved by
increases in the area of sensitive tissue within each eye. In insects, it seems likely that
increased sensitivity was achieved by multiplication of the number of small photore-
ceptors, each in its own cup. At this early stage, these alternatives yielded equally use-
ful eyes. Once the insect lineage had evolved a visual system based on many small eyes,
however, images could not be formed the same way as in camera-type eyes because
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 80
Compound
Camera-type eye
eye
Researchers believe that differ-
FIGURE 3.24
ent evolutionary pathways led to
the development of camera-type and compound eyes.
Selection for greater light sensitivity in a simple organism More eyespots
Larger eyespot
could favor the development of a larger eyespot or of mul-
tiple eyespots. The first pathway could lead to camera-
type eyes, and the second to compound eyes. Images are
formed very differently in compound and camera-type
eyes, making it virtually impossible for compound eyes to
evolve into camera-type eyes. Eyespot
whereas the camera-type eye inverts images, the compound eye does not. It does not
seem structurally possible to evolve from a compound eye to a camera-type eye in a
series of small steps, each favored by selection.
Box 3.2
The Geometry of Area/
Volume Ratios
When an animal becomes larger without otherwise chang- to volume is halved. We will see later that this fact has con-
ing shape, the ratio of any fixed area measurement to its vol- sequences for how temperature affects the size of animals.
ume decreases. This is easiest to understand if we compute The same geometric principle governs the relationship
how the ratio of an animal’s entire surface area to its volume between any area measure and volume. Suppose the cubic
changes as the animal becomes larger. To see why, suppose animal has a vertical bone running through its center that
that the animal is a cube x centimeters on a side. Then its supports the weight of the rest of the animal. For simplicity,
volume is x3 and its surface area is 6x2. Thus the ratio of its we make the bone square in cross section and assume that
surface area to its volume is it has the dimensions ᎐12 x by ᎐12 x, as shown in Figure 3.25.
The cross-sectional area of this bone, then, is ᎐12 x ✕ ᎐12 x =
surface area 6x2 6 ᎐41 x2. Thus, the ratio of the cross-sectional area of the bone
5 5
volume x 3
x to the volume of the animal is
1
This means that an animal measuring 1 cm on a side has 6 4 x2 1
cm2 of surface area for each cubic centimeter of volume. An 5
x3 4x
animal 2 cm on a side has only 3 cm2 of surface area for
each cubic centimeter of volume. Of course, there aren’t any So if the linear dimensions of an animal are doubled, its
cubic animals, but it turns out that the shape doesn’t alter weight will increase eightfold, but the cross-sectional area of
this relationship. When an animal’s linear dimension is its bones—and therefore their strength—will increase only
doubled without a change in shape, the ratio of surface area fourfold.
Bone
x
x
1x 1x
2 2
81
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 82
FURTHER READING
Study Questions 83
and the sickle-cell allele is designated S. There are three hemoglobin genotypes,
and the phenotypes associated with these genotypes can be distinguished in a vari-
ety of ways. In one African population of 10,000 adults, for example, there are
3,000 AS individuals, 7,000 AA individuals, and 0 SS individuals.
(a) Suppose that these individuals were to mate at random. What would be the
frequency of the A and S alleles among the gametes that they produced?
(b) What would be the frequency of the three genotypes that resulted from ran-
dom mating?
(c) Is the original population of adults in Hardy–Weinberg equilibrium?
2. The three common genotypes at the hemoglobin locus have very different phe-
notypes: SS individuals suffer from severe anemia, AS individuals have a relatively
mild form of anemia but are resistant to malaria, and AA individuals have no ane-
mia but are susceptible to malaria. The frequency of the S allele among the
gametes produced by the first generation of a central African population is 0.2.
(a) Assuming that mating occurs at random, what are the frequencies of the three
genotypes among zygotes produced by this population?
(b) In this area, no SS individuals survive to adulthood, 70% of the AA individ-
uals survive, and all of the AS individuals survive. What is the frequency of
each of the three genotypes among the second generation of adults?
(c) What is the frequency of the S allele among gametes produced by these adults?
3. Tay-Sachs disease is a lethal genetic disease controlled by two alleles: the Tay-
Sachs allele T and the normal allele N. Children who have the TT genotype suf-
fer from mental deterioration, blindness, paralysis, and convulsions; and they die
sometime between the ages of three and five years. The proportion of infants
afflicted with Tay-Sachs disease varies in different ethnic groups. The frequency
of the Tay-Sachs allele is highest among descendants of central European Jews.
Suppose that in a population of 5,000 people descended from central European
Jews the frequency of the Tay-Sachs allele is 0.02.
(a) Assuming that members of this population mate at random and each adult has
four children on average, how many gametes from each person are success-
ful at being included in a zygote?
(b) Calculate the frequency of each of the three genotypes among the population
of zygotes.
(c) Assume that no TT zygotes survive to become adults, and that 50% of the TN
and NN individuals survive to become adults. What is the frequency of the
TT and TN genotypes among these adults?
(d) When the adult individuals resulting from zygotes in (b) produce gametes,
what is the frequency of the T allele among their gametes? Compare your
answer to the frequency of the T allele among gametes produced by their par-
ents. Think about your result, given that these numbers are roughly realistic
and that there have been no substantial medical breakthroughs in the treat-
ment of Tay-Sachs disease. What is the paradox here?
4. Consider a hypothetical allele that is lethal, like the Tay-Sachs allele, but is dom-
inant rather than recessive. Assume that this allele has a frequency of 0.02, like
the Tay-Sachs allele. Recalculate the answers to (a), (b), and (c) of question 3. Is
selection stronger against the recessive allele or the dominant one? Why? Assum-
ing that mutation to the deleterious allele occurs at the same rate at both loci,
which allele will occur at a higher frequency? Why?
5. Consider the Rh blood group. For this example we will assume that there are only
two alleles, R and r, of which r is recessive. Homozygous rr individuals produce
certain nonfunctional proteins and are said to be Rh-negative (Rh–), while Rr and
03_Modern Synthesis [4p]:How Humans Evolved [5e] 10/10/08 11:22 AM Page 84
RR individuals are Rh-positive (Rh+). When an Rh– female mates with an Rh+
male, their offspring may suffer serious anemia while still in the uterus. The fre-
quency of the r allele in a hypothetical population is 0.25. Assuming that the pop-
ulation is in Hardy–Weinberg equilibrium, what fraction of the offspring in each
generation will be at risk for this form of anemia?
6. The blending model of inheritance was attractive to nineteenth-century biologists
because it explained why offspring tend to be intermediate in appearance between
their parents. This model fails, however, because even though it predicts a loss in
variation, variation is maintained. How does Mendelian genetics explain the inter-
mediate appearance of offspring without the loss of variation? Be sure to explain
why the properties of the Hardy–Weinberg equilibrium are an important part of
this explanation.
7. In a particular species of fish, egg size and egg number are negatively correlated.
Draw a graph that illustrates this fact. What will happen to egg size if selection
favors larger numbers of eggs?
8. A plant breeder trying to increase the yield of wheat selects plants with larger ker-
nels. After several generations of selection, the size of kernels produced has
increased substantially. However, the number of kernels per plant has decreased,
so the total yield remains constant. What two sources of maladaptation best
explain this result?
9. The compound eyes of insects provide poor image clarity compared with the
images produced by the camera-type eyes of vertebrates. Explain why insects don’t
evolve camera-type eyes.
10. Explain why genetic drift has no effect on genetic loci that are not variable. (When
all individuals in the population are homozygous for the same allele, that genetic
locus is not variable.) What does this mean about the long-term outcome of
genetic drift?
11. Why does natural selection produce adaptations only at equilibrium?
12. Two dog breeders are using artificial selection to create new breeds. Both begin
with dachshunds. One breeder wants to create a breed with longer forelimbs and
hindlimbs. The second breeder wants to create a breed with shorter forelimbs and
longer hindlimbs. Which breeder will make more rapid progress? Why?
13. When selection makes animals larger overall, their bones also get thicker. The
increase in bone thickness could be due to the correlated response to selection, or
to independent selection for thicker bones, or to some combination of the two fac-
tors. How could you determine the relative importance of these two processes?
KEY TERMS
population genetics correlated response (positive and
genotypic frequency negative correlation)
gene frequency pleiotropic effects
Hardy–Weinberg equilibrium maladaptive
modern synthesis sampling variation
environmental variation genetic drift
mutations fixation
mate guarding camera-type eyes
sex ratio compound eyes
canalized development
plastic lactate
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 85
Chapter 4
What Are Species?
The Biological Species Concept
The Ecological Species Concept
The Origin of Species
Allopatric Speciation
Parapatric and Sympatric Speciation
The Tree of Life
Why Reconstruct Phylogenies?
How to Reconstruct Phylogenies
Problems Due to Convergence
Problems Due to Ancestral Characters
Using Genetic-Distance Data to
Date Phylogenetic Events
Taxonomy: Naming Names
S
o far, we have focused on how natural selection, mutation, and genetic drift
cause populations to change through time. These are mechanisms of microevo-
lution, and they affect the morphology, physiology, and behavior of particular
species in particular environments. For example, microevolutionary processes are
responsible for variation in the size and shape of the beaks of medium ground finches
on Daphne Major.
85
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 86
However, this is not all there is to the process of evolution. Darwin’s major work
was entitled On the Origin of Species because he was interested in how new species
are created, as well as how natural selection operates within populations. Evolution-
ary theory tells us how new species, genera, families, and higher groupings come into
existence. These processes are mechanisms of macroevolution. Macroevolutionary
processes play an important role in the story of human evolution. To properly inter-
pret the fossil record and reconstruct the history of the human lineage, we need to
understand how new species and higher groupings are created and transformed over
time.
Organisms cluster into distinct types called species. The individual organisms that
belong to a species are similar to each other and are usually quite distinct from the
members of other species. For example, in Africa some tropical forests house two
species of apes: chimpanzees and gorillas. These two species are similar in many ways:
both are tailless, both bear weight on their knuckles when they walk, and both defend
territories. Nonetheless, these two species can easily be distinguished on the basis of
their morphology: chimpanzees are smaller than gorillas; male gorillas have tiny testes
(singular testis; the organs that produce sperm), while male chimpanzees have quite
large ones; and gorillas have a fin of bone on their skull, while chimpanzees have more
rounded skulls. Gorillas and chimpanzees also differ in their behavior: chimpanzees
make and use tools in foraging, while gorillas do not; male gorillas beat their chests
when they perform displays, while male chimpanzees flail branches and charge about;
and gorillas live in smaller groups than chimpanzees do. These two species are easy
to distinguish because no animals are intermediate between them; there are no
“gimps” or “chorillas” (Figure 4.1).
Species are not abstractions created by scientists; they are real biological cate-
gories. People all over the world name the plants and animals around them, and biol-
ogists use the same kinds of phenotypic characteristics to sort animals into species that
other people use. For the most part, there is little problem in identifying any particu-
lar specimen from its phenotype.
Although nearly everyone agrees that species exist and can recognize species in
nature, biologists are much less certain about how species should be defined. This
Chimpanzees (a)
FIGURE 4.1
and gorillas (b)
sometimes occupy the same forests and
share certain traits. However, these two
species are readily distinguished because
no animals are intermediate between
them.
(a) (b)
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 87
uncertainty arises from the fact that evolutionary biologists do not agree about why
species exist. There is now a considerable amount of controversy about the processes
that give rise to new species and the processes that maintain established ones.
Although there are many different views on these topics, we will concentrate on two
of the most widely held points of view: the biological species concept and the ecolog-
ical species concept.
Most zoologists believe in the biological species concept, which defines a biological
species as a group of organisms that interbreed in nature and are reproductively iso-
lated. Reproductive isolation means that members of a given group of organisms do
not mate successfully with organisms outside the group. For example, there is just one
species of gorilla, Gorilla gorilla, and this means that all gorillas are capable of mat-
ing with one another and they do not breed with any other kinds of animals in nature.
According to adherents of the biological species concept, reproductive isolation is the
reason why there are no gimps or chorillas.
The biological species concept defines a species in terms of the ability to interbreed
because successful mating leads to gene flow, the movement of genetic material within
parts of a population or from one population to another. Gene flow tends to maintain
similarities among members of the same species. To see how gene flow preserves homo-
geneity within species, consider the hypothetical situation diagrammed in Figure 4.2a.
Imagine a population of finches living on a small island in which there are two habi-
tats: wet and dry. Natural selection favors different-sized beaks in each habitat. Large
beaks are favored in the dry habitat, and small beaks are favored in the wet habitat.
Because the island is small, however, birds fly back and forth between the two envi-
ronments and mate at random, so there is a lot of gene flow between habitats. Unless
selection is very strong, gene flow will swamp its effects. On average, birds in both
habitats will have medium-sized beaks, a compromise between the optimal phenotypes
for each habitat. In this way, gene flow tends to make the members of a species evolve
as a unit.
Now suppose that there are finches living on two different islands, one wet and
one dry, and that the islands are far enough apart that the birds are unable to fly from
one island to the other (Figure 4.2b). This means that there will be no interbreeding
and no gene flow will occur. With no genetic exchange between the two independent
groups to counter the effects of selection, the two populations will diverge genetically
and become less similar phenotypically. Birds on the dry island will develop large
beaks, and birds on the wet island will develop small beaks.
88
Fitness in each habitat C H A P T E R 4 • Speciation and Phylogeny
(a) (b)
Gene flow among populations destroys differences between them. (a) Suppose a
FIGURE 4.2
population of finches lives on an island with both wet and dry habitats. In the dry
habitat, selection favors deep beaks, but selection in the wet habitat favors shallow beaks. Because
interbreeding leads to extensive gene flow between populations in the two habitats, the population
beak size responds to an average of the two environments. (b) Now suppose two populations of finches
live on different islands, one wet and one dry, with no gene flow between the two islands. In this sce-
nario, deep beaks are common on the dry island, and shallow beaks are common on the wet island.
species would merge into one. There are no chorillas or gimps in nature because chim-
panzees and gorillas are reproductively isolated.
Reproduction is a complicated process, and anything that alters the process can
act as an isolating mechanism. Even subtle differences in activity patterns, courtship
behavior, or appearance may prevent individuals of different types from mating. More-
over, even if a mating among individuals of different types does take place, the egg may
not be fertilized or the zygote may not survive.
Critics of the biological species concept point out that gene flow is neither necessary
nor sufficient to maintain species boundaries in every case, and they argue that selec-
tion plays an important role in preserving the boundaries between species. The view
that emphasizes the role of natural selection in creating and maintaining species is
called the ecological species concept.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 89
Medium- Medium-
small large
hybrid hybrid In the Galápagos, selection main-
FIGURE 4.3
tains three species of ground
finches, even though there is substantial gene flow among
them. The red line represents the amount of food avail-
able in the environment for birds with different-sized
beaks. Each peak in this curve represents a different
Beak depth species.
In nature, species boundaries are often maintained even when there are substan-
tial amounts of gene flow between species. For example, the medium ground finch
readily breeds with the large ground finch on islands where they coexist. Peter Grant
and his colleagues have estimated that approximately 10% of the time, medium
ground finches mate with large ground finches, leading to considerable gene flow
between these two species. Yet they have not merged into a single species. Grant and
his colleagues have concluded that the medium ground finch and the large ground
finch have remained distinct because these two species represent two of the three opti-
mal beak sizes for ground finches (Figure 4.3). These three optimal sizes are based on
the availability of seeds of different size and hardness and the ability of birds with
different-sized beaks to harvest these seeds. According to the calculations of Grant’s
group, the three optimal beak sizes for ground finches correspond to the average beak
sizes of the small (Geospiza fuliginosa), medium (G. fortis), and large (G. magnirostris)
ground finch species (Figure 4.4). These researchers suggest that hybrids are selected
against because their beaks fall in the “valleys” between these selective “peaks.” The
kind of interbreeding seen in the Galápagos finches is not particularly unusual. A sur-
vey of 114 plant species and 170 animal species by Loren Rieseberg and his collabo-
(a) The small ground finch, Geospiza fuliginosa; (b) the medium ground finch,
FIGURE 4.4
G. fortis; (c) the large ground finch, G. magnirostris.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 90
Amphibians
Birds
Fishes
Ferns
Insects
Mammals
rators at Indiana University indicates that a substantial fraction of species are not
reproductively isolated (Figure 4.5).
In addition, a number of species have maintained their coherence with no gene
flow between isolated subpopulations. For example, the checkerspot butterfly (Figure
4.6) is found in scattered populations throughout California. Members of different
populations are very similar morphologically and are all classified as members of the
species Euphydryas editha. However, careful studies by Stanford University biologist
Paul Ehrlich have shown that these butterflies rarely disperse
more than 100 m (about 110 yds) from their place of birth.
Given that populations are often separated by several kilo-
meters and sometimes by as much as 200 km (about 124
miles), it seems unlikely that there is enough gene flow to
unify the species.
The existence of strictly asexual organisms provides
additional evidence that species can be maintained without
gene flow. Asexual species, which reproduce by budding or
fission, simply replicate their own genetic material and pro-
duce exact copies of themselves. Thus, gene flow cannot
occur in asexual organisms, and it does not make sense to
think of them as being reproductively isolated. Nonetheless,
many biologists who study purely asexual organisms con-
Checkerspot butterflies living in localized popu- tend that it is just as easy to classify them into species as to
FIGURE 4.6 classify organisms that reproduce sexually.
lations throughout California are all members of
the same species, but there is probably very little gene flow among What maintains the coherence of species in these cases?
different populations. Most biologists think the answer is natural selection. To see
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 91
why, let’s return to the Galápagos once more. Imagine a situation in which selection
favored finches with small beaks in one habitat and finches with large beaks in another
habitat, but did not favor birds with medium-sized beaks in either habitat. If birds with
medium-sized beaks consistently failed to survive or reproduce successfully, natural
selection could maintain the difference between the two bird species even if they inter-
bred freely. In the case of the checkerspot butterflies and asexual species, we can imag-
ine the opposite scenario: selection favors organisms with the same morphology and
physiology and thus maintains their similarity even in the absence of gene flow.
Today most biologists concede that selection maintains species boundaries in a few
odd cases like Darwin’s finches, but they generally insist that reproductive isolation
plays the major role in most instances. However, a growing minority of researchers
contend that selection plays an important role in maintaining virtually all species
boundaries.
The species is one of the most important concepts in biology, so it would be helpful
to know how new species come into existence. Despite huge amounts of hard work
and many heated arguments, however, there is still uncertainty about what Darwin
called the “mystery of mysteries”—the origin of species. The mystery persists because
the process of speciation is very difficult to study empirically. Unlike microevolution-
ary change within populations, which researchers are sometimes able to study in the
field or laboratory, new species usually evolve too slowly for any single individual to
study the entire process. And on the flip side, speciation usually occurs much too rap-
idly to be detected in the fossil record. Nonetheless, biologists have compiled a sub-
stantial body of evidence that provides important clues about the processes that give
rise to new species.
Allopatric Speciation
Colonization of
isolated island
Wet
Dry
Dry
tion from some other small bird adapted to the dry habitat and there is very little
movement between the two islands, the population of finches on the dry island will
rapidly adapt to their new habitat and the birds’ average beak size will increase (Fig-
ure 4.7).
Now let’s suppose that after some time, finches from the small, dry island are
blown back to the large island from which their ancestors came. If the large-beaked
newcomers successfully mate with the medium-beaked residents, then gene flow
between the two populations will rapidly eliminate the differences in beak size
between them, and the recently created large-beaked variety will disappear. If, on the
other hand, large-beaked immigrants and small-beaked residents cannot successfully
interbreed, then the differences between the two populations will persist. As we noted
earlier, a variety of mechanisms can prevent successful interbreeding, but it seems that
the most common obstacle to interbreeding is that hybrid progeny are less viable than
other offspring. In this case, we might imagine that when the two populations of
finches became isolated, they diverged genetically because of natural selection and
genetic drift. The longer they remain isolated, the greater the genetic difference
between the populations becomes. When the two distinct types then come into con-
tact, hybrids may have reduced viability, either because genetic incompatibilities have
arisen during their isolation or because hybrid birds are unable to compete successfully
for food. If these processes cause complete reproductive isolation, then a new species
has been formed.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 93
The distribution of beak sizes of the small ground finch (G. fuliginosa) and the
FIGURE 4.8
medium ground finch (G. fortis) on three of the Galápagos Islands illustrates the
effects of character displacement. (b) The distribution of beak depth for each species on Santa Cruz Daphne:
Only medium
Island, where the two species compete. (a, c) The distributions of beak depth for the small and medium ground finch
Frequency
ground finches on islands where they do not compete with each other (G. fortis on Daphne Major and
G. fuliginosa on Los Hermanos). The beaks of the two species are the most different where there is
direct competition. Careful measurements of seed density and other environmental conditions suggest
that competition, not environmental differences, is responsible for the difference between the
populations.
5 10 15
(a)
Even if there is some gene flow after the members of the two populations initially
come back into contact, two additional processes may increase the degree of repro-
Santa Cruz:
ductive isolation and facilitate the formation of a new species. The first process, char- Medium
acter displacement, may occur if competition over food, mates, or other resources Small ground
Frequency
increases the morphological differences between the immigrants and the residents. ground finch
finch
In our example, the large-beaked immigrants will be better suited to the dry parts of
the large island and will be able to outcompete the original residents of these areas. Res-
ident birds will be better off in the wet habitats, where they face less competition from
the immigrants. Because small beaks are advantageous in wet habitats, residents with
5 10 15
smaller-than-average beaks will be favored by selection. At the same time, because large (b)
beaks are advantageous in dry habitats, natural selection will favor increased beak size
among the immigrants. This process will cause the beaks of the competing populations
to diverge. There is good evidence that character displacement has played an impor-
tant role in shaping the morphology of Darwin’s finches (Figure 4.8). Los Hermanos:
Only small
A second process, called reinforcement, may act to reduce the extent of gene flow ground finch
Frequency
between the populations. Because hybrids have reduced viability, selection will favor
behavioral or morphological adaptations that prevent matings between members of
the two populations. This process will further increase the reproductive isolation
between the two populations (Figure 4.9). Thus, character displacement and rein-
forcement may amplify the initial differences between the populations and lead to two
new species. 5 10 15
Beak depth (mm)
(c)
Allopatric speciation requires a physical barrier that initially isolates part of a pop-
ulation, interrupts gene flow, and allows the isolated subpopulation to diverge from
the original population under the influence of natural selection. In our example, the
physical barrier is the sea, but mountains, rivers, and deserts can also restrict move-
ment and interrupt gene flow. Character displacement and reinforcement may work
to increase differences when members of the two populations renew contact. However,
these processes are not necessary elements in allopatric speciation. Many species
become completely isolated while they are separated by a physical barrier.
(a) (b)
(c) (d)
Baboons are distributed all over Africa and occupy a very diverse range of habi-
FIGURE 4.10
tats. (a) In the Drakensberg Mountains in South Africa, winter temperatures
drop below freezing, and snow sometimes falls. (b) In the Matopo Hills of Zimbabwe, woodlands are
interspersed with open savanna areas. (c) Amboseli National Park lies at the foot of Mount Kilimanjaro
just inside the Kenyan border. (d) Gombe Stream National Park lies on the hilly shores of Lake
Tanganyika in Tanzania.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 95
New species may also form if there is strong selection that favors two different
phenotypes.
Biologists who endorse the biological species concept usually argue that allopatric spe-
ciation is the most important mechanism for creating new species in nature. For these
scientists, gene flow “welds” a species together, and species can be split only if gene flow
is interrupted. Biologists who think that natural selection plays an important role in
maintaining species often contend that selection can lead to speciation even when there
is interbreeding. There are strong and weak versions of this hypothesis. The weak ver-
sion, called parapatric speciation, holds that selection alone is not sufficient to produce
a new species, but new species can be formed if selection is combined with partial genetic
isolation. For example, baboons range from Saudi Arabia to the Cape of Good Hope
and occupy an extremely broad range of environments. Some baboons live in moist
tropical forests, some live in arid deserts, and some live in high-altitude grasslands (Fig-
ure 4.10). Different behaviors and morphological traits may be favored in each of these
environments, and this variation may cause baboons in different regions to vary. At
habitat boundaries, animals that come from different habitats and have different char-
acteristics may mate and create a hybrid zone. Study of hybrid zones in a wide variety
of species suggests that hybrids are usually less fit than nonhybrids. When this is the case,
selection should favor behavior or morphology that prevents mating between members
of individuals from different habitats. If such reinforcement does occur, gene flow will
be reduced, and eventually two new, reproductively isolated species will evolve.
The strong version of the hypothesis, called sympatric speciation, contends that
strong selection favoring different phenotypes can lead to speciation even when there
is no geographic separation, and therefore initially there is extensive gene flow among
individuals in the population. Sympatric speciation is theoretically possible, and this
form of speciation has been induced in laboratory populations, but it is uncertain
whether it occurs in nature. The three speciation mechanisms are diagrammed in Fig-
ure 4.11.
When the first finches arrived in the Galápagos 500,000 years ago, there
FIGURE 4.12
were many empty niches. Some finches became cactus eaters (a), others
became seed eaters (b), and some became predators on insects and other arthropods (c).
Ecologists use the term niche to refer to a particular way of “making a living,”
which includes the kinds of food eaten, as well as when, how, and where the food is
acquired. One consequence of all three models of speciation is that the rate of speci-
ation depends on the number of available ecological niches. Once again, Darwin’s
finches provide a good example. When the first finches migrated to the Galápagos from
the mainland of South America (or perhaps from the Cocos Islands) about half a mil-
lion years ago, all of the niches for small birds in the Galápagos were empty. There
were opportunities to make a living as a seed eater, a cactus eater, and so on. The
finches’ ancestors diversified to fill all of these ecological niches, and eventually they
became 14 distinct species (Figure 4.12). An even more spectacular example of this
process occurred at the end of the Cretaceous era. Dinosaurs dominated the Earth dur-
ing the Cretaceous era but disappeared suddenly 65 mya (million years ago). The
mammals that coexisted with the dinosaurs were mostly small, nocturnal, and insec-
tivorous. But when the dinosaurs became extinct, these small creatures diversified to
fill a broad range of ecological niches, evolving into elephants, killer whales, buffalo,
wolves, bats, gorillas, humans, and other kinds of mammals. When a single kind of
Darwin’s finches animal or plant diversifies to fill many available niches, the process is called adaptive
FIGURE 4.13
are not the only radiation (Figure 4.13).
example of adaptive radiations that
occur when immigrants encounter a
range of empty ecological niches. In the
Hawaiian archipelago, the adaptive radia- THE TREE OF LIFE
tion of honeycreeper finches, shown
here, produced an even greater diversity
of species.
Organisms can be classified hierarchically on the basis of similarities. Many such
similarities are unrelated to adaptation.
(a)
opponents in the debates following the publication of On the Origin of Species, used
dugongs (aquatic mammals much like manatees), bats, and moles as an example of this
phenomenon. All three of these creatures have the same kind and number of bones in
their forelimbs, even though the shapes of these bones are quite different (Figure 4.14).
The forelimb of the bat is adapted to flying, and that of the dugong to paddling (Fig-
ure 4.15). Nonetheless, the basic structure of a bat’s forelimb is much more similar to
that of a dugong than to the forelimb of a swift, even though the swift’s forelimb is
also designed for flight.
Such patterns of similarity make it possible to cluster
species hierarchically—like a series of nested boxes (Figure
4.16). This remarkable property of life is the basis of the sys-
tem for classifying plants and animals devised by the
eighteenth-century Swedish biologist Carolus Linnaeus. All
species of bats share many similarities and are grouped
together in one box. Bats can be clustered with dugongs and
moles in a larger box that contains all of the mammals.
Mammals in turn are classified together with birds, reptiles,
and amphibians in an even larger box. Sometimes the simi-
larities that lead us to classify animals together are functional
similarities. Many of the features shared by different species
of bats are related to the fact that they fly at night. However,
many shared features bear little relation to adaptation; bats,
tiny aerial acrobats, for example, are grouped with the enor- The dugong is an aquatic animal. Its forelimb
FIGURE 4.15
mous placid dugong, and not with the small acrobatic swift. is adapted for paddling through the water.
The fact that new species derive from existing species accounts for the existence
of the patterns of nonadaptive similarity that allow organisms to be classified hier-
archically. Clearly, new species originate by splitting off from older ones, and we can
arrange a group of species that share a common ancestor into a family tree, or
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 98
Vertebrates
Mammals Birds
Dugongs Swifts
Bats Penguins
Moles Kiwis
Patterns of similarity allow organisms to be
FIGURE 4.16
classified hierarchically into a series of nested
boxes. All mammals share more similarities with each other than they
do with birds, even though some mammals, such as bats, must solve the
same adaptive problems that face birds.
phylogeny. Figure 4.17 shows the family tree of the hominoids—the superfamily that
includes apes and humans—and Figure 4.18 shows what some of these present-day
apes look like. At the root of the tree is an unknown ancestral species from which all
hominoids evolved. Each branch represents a speciation event in which one species
split into two daughter species, and one or both of the new daughter species diverged
in morphology or behavior from the parent species.
It is important to realize that when two daughter species diverge from each other,
they don’t differ in all phenotypic details. A few traits may differ while most others
retain their original form. For example, the hominoids share many common features,
including an unspecialized digestive system, five toes on each foot, and no tail. At the
same time, these animals have diverged in a number of ways. Gorillas live in groups
Siamang
Gibbon
Orangutan
Gorilla
Human
Gibbons (a) and siamangs are called “lesser apes” and are classified in the fam-
FIGURE 4.18
ily Hylobatidae. Orangutans (b), gorillas (c), and chimpanzees (d) are called
“great apes” and are classified in the family Pongidae.
that contain one adult male and several adult females; orangutans are mainly solitary.
Male and female gibbons are about the same size, but among the other apes, males are
larger than females. In general, we expect to see the greatest divergence in those traits
that are related to making a living in different habitats and to choosing mates.
Now let’s return to the family tree. Each time one species splits to become two new
species, the new daughter species will differ in some way. They will continue to diverge
through time because once speciation has occurred, the two lineages evolve inde-
pendently. Some differences will arise because the two species adapt to different habi-
tats; other differences may result from random processes like genetic drift. In general,
species that have recently diverged will have more characteristics in common with one
another than with species that diverged in the more distant past. For example, chim-
panzees and gorillas share more traits with one another than they do with orangutans
or gibbons because they share a more recent common ancestor. This pattern of shar-
ing traits is the source of the hierarchical nature of life.
We have seen that descent with modification explains the hierarchical structure of the
living world. Because new species always evolve from existing species and species are
reproductively isolated, all living organisms can be placed on a single phylogenetic tree,
which we can then use to trace the ancestry of all living species. In the remainder of
this chapter, we will see how the pattern of similarities and differences observed in liv-
ing things can be used to construct phylogenies and to help establish the evolutionary
history of life.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 100
Bipedal
walking Humans
Non- evolves Bipedal
knuckle- walking
walking
Non- ancestor Knuckle
knuckle- walking Chimpanzee
walking evolves Knuckle
ancestor walking
Knuckle
walking Gorilla
evolves Knuckle
walking
(b) Knuckle walking evolved independently in chimpanzees and gorillas
ferent primate species. However, most biologists believe that only independently
evolved cases should be counted in comparative analyses, so we must take the
phylogenetic relationships among species into account. Box 4.1 provides a
hypothetical example of how phylogenetic information can alter our interpre-
tations of comparative data.
For many years, scientists constructed phylogenies only for the purpose of classi-
fication. The terms taxonomy and systematics were used interchangeably to refer to
the construction of phylogenies and to the use of such phylogenies for naming and clas-
sifying organisms. With the recent realization that phylogenies have other important
uses like the ones just described, there is a need for terms that distinguish phylogenetic
construction from classification. Here we adopt the suggestion of University of Chicago
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 102
Box 4.1
The Role of Phylogeny in
the Comparative Method
To understand why it is important to take phylogeny into terrestriality are found together only in species B and its
account when using the comparative method, consider the descendants: B1, B2, and B3. This combination evolved only
phylogeny shown in Figure 4.22, which shows the pattern once, although we now observe this combination in three
of relationships for eight hypothetical primate species. living species. There is also only one case of selection creat-
As you can see, three terrestrial species live in large ing an arboreal species that lives in small groups (species C
groups and three arboreal species live in small groups. Only and its descendants: C1, C2, and C3). Note that each of the
one terrestrial species lives in small groups, and only one other possible combinations has also evolved once. When
arboreal species lives in large groups. Thus, if we based our we tabulate independent evolutionary events, we find no
determinations on living species, we would conclude that consistent relationship between lifestyle and group size.
there is a statistical relationship between group size and Clearly, phylogenetic information is crucial to making sense
lifestyle. If we count independent evolutionary events, how- of the patterns we see in nature.
ever, we get a very different answer. Large group size and
X small,
arboreal
102
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 103
TABLE 4.1 The amino acid sequence for the protein myoglobin is shown for three differ-
ent species. The numbers refer to positions on the myoglobin chain, and the
letter in each cell stands for the particular amino acid found at that position in each kind of myoglo-
bin. All three species have the same amino acids at all of the positions not shown here, as well as at
positions 1 to 3. At eight positions, however, there is at least one discrepancy among the three
species (shaded).
AMINO ACID NUMBER
Species 1 2 3 5 9 13 30 34 48 59 66
A G L S G L V I K H E A
B G L S G L V I K G A I
C G L S Q Q I M H G E Q
anthropologist Robert Martin to employ the term systematics to refer to the con- A B C
struction of phylogenies, and the term taxonomy to mean the use of phylogenies in
naming and classification. Although this distinction may not seem important now, it
will become more relevant as we proceed.
humans and duck-billed platypuses (A and B) are more closely related to each other
than either of them is to chickens (C). However, this phylogeny is not based on nearly
enough data to be conclusive. To be convinced of the relationships among these three
organisms, we would need to gather and analyze data on many more characters and
generate the same tree each time. In fact, many other characters show the same pat-
tern as myoglobin. Humans and duck-billed platypuses, for example, have hair and
mammary glands, but chickens lack these structures. The phylogeny in Figure 4.23 is
the currently accepted tree for these three species.
Despite the evidence just discussed, not everything about the phylogeny for humans,
platypuses, and chickens is hunky-dory. Two amino acid positions, 48 and 59 (see
Table 4.1), are not consistent with the phylogeny shown in Figure 4.23. Moreover,
certain other characters don’t fit this tree. For example, both platypuses and humans
lactate, but chickens don’t (Figure 4.25a); both humans and chickens are bipedal, but
platypuses are not (Figure 4.25b); and both platypuses and chickens lay eggs (Figure
4.25c), have a feature of the gut called a cloaca, and sport horny bills, but humans
don’t. Why don’t these characters fit neatly into our phylogeny?
One reason for these anomalies is convergent evolution. Sometimes traits shared
by two species are not the result of common ancestry. Instead, they are separate adap-
tations independently produced by natural selection. Chickens and humans are bipedal
not because they are descended from the same bipedal ancestor, but rather because
they each evolved this mode of locomotion independently. Similarly, the horny bills of
chickens and platypuses are not similarities due to descent; they are independently
Lactates Bipedal
No eggs
Doesn't Eggs
lactate
Quadrupedal
(a) Many traits, such as lactation, generate the same pattern of relationships
FIGURE 4.25
among humans, chickens, and platypuses that myoglobin does (see Figure
4.23). (b) Other characters, such as bipedal locomotion, suggest a closer relationship between humans
and chickens than between humans and platypuses or between platypuses and chickens. (c) Some
traits, like egg laying, suggest a closer relationship between chickens and platypuses than between
chickens and humans or between platypuses and humans.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 105
derived characters. Systematists say that characters similar because of convergence are
analogous, while characters whose similarity is due to descent from a common ances-
tor are homologous. It is important to avoid using convergent traits in reconstructing
phylogenetic relationships. Although it is fairly obvious that bipedal locomotion in
humans and chickens is not homologous, convergence is sometimes very difficult to
detect.
There are also homologous traits that do not fit neatly into correct phylogenies. For
example, chickens and platypuses reproduce by laying eggs, while humans do not. It
seems likely that both chickens and platypuses lay eggs because they are descended
from a common egg-laying ancestor. But if these characters are homologous, why don’t
they allow us to generate the correct phylogeny (see Figure 4.25c)?
Egg laying is an example of what systematists call an ancestral trait, one that char-
acterized the common ancestor of chickens, platypuses, and humans. Egg laying has
been retained in the chicken and the platypus but lost in humans. It is important to
avoid using ancestral characters when constructing phylogenies. Only derived traits—
features that have evolved since the time of the last common ancestor of the species
under consideration—can be used in constructing phylogenies.
To see why we need to distinguish between ancestral and derived traits, consider
the three hypothetical species of “cooties” pictured in Figure 4.26. At first glance the
red-eyed cootie and the blue-necked cootie seem more similar to one another than
either is to the orange-spotted cootie, and a careful count of characters will show that
they do share more traits with one another than with the orange-spotted variety. But
consider Figure 4.27, which shows the phylogeny for the three species. You can see
that the blue-necked cootie is actually more closely related to the orange-spotted cootie
because they share a more recent common ancestor. The blue-necked cootie seems
more similar to the red-eyed cootie because they share many ancestral characters, but
the orange-spotted cootie has undergone a period of rapid evolution that has elimi-
nated most ancestral characters. Looking at ancestral traits will not generate the cor-
rect phylogeny if rates of evolution differ among species.
If we base our assessment of similarity only on the number of derived characters
that each species displays (as shown in Figure 4.28), then the most similar species
are the ones most closely related. The blue-necked cootie and orange-spotted cootie
share one derived character (an orange face) with each other, but they share no derived
characters with the red-eyed cootie. Thus, if we avoid ancestral characters, we can con-
struct the correct phylogeny.
106
genetic-distance estimates based on DNA sequence data. For example, to compute the
genetic distance between humans and chimpanzees, the biologist first identifies homol-
ogous DNA segments in the two species. This means that the segments are descended
from the same DNA sequence in the common ancestor of humans and chimpanzees.
These DNA segments are then sequenced. The number of nucleotide sites at which the
two sequences differ is used to compute the genetic distance between the two species.
The mathematical formulas used for these calculations are beyond the scope of this
book, but suffice it to say that the larger the number of nucleotide differences, the big-
ger the genetic distance.
Genetic-distance data are often consistent with the hypothesis that genetic
distance changes at an approximately constant rate.
The genetic distances among noncoding sequences of the DNA of humans, chim-
panzees, gorillas, and orangutans are shown in Table 4.2. Notice that gorillas,
humans, and chimpanzees are essentially the same genetic distance from orangutans.
This close similarity is evidence that genetic distance changes at an approximately con-
stant rate. To see why, think of genetic distance accumulating along the path leading
from the last common ancestor of all of the hominoids to each living species. The total
distance between any pair of living species is the sum of their paths from the last com-
mon ancestor. For example, the genetic distance between humans and orangutans is
the accumulated distance between the last common ancestor of all four of these species
and orangutans plus the accumulated distance between the last common ancestor and
humans. Now notice that the distance between orangutans and humans is the same
as the distance between chimpanzees and orangutans. This means that the genetic dis-
tance between the last common ancestor of orangutans and humans is the same as the
distance between the last common ancestor of chimpanzees and orangutans. Because
the time that elapsed since the branching off from the last common ancestor is the
same for all these species, it follows that the rate of change of genetic distance along
both of these paths through the phylogeny must be approximately the same, and that
the genetic distance between two species is a measure of the time elapsed since both
had a common ancestor. Evolutionists refer to genetic distances with a constant rate
of change as molecular clocks because genetic change acts like a clock that measures
the time since two species shared a common ancestor. Data from many other groups
of organisms suggest that genetic distances often have this clocklike property, but there
are also important exceptions.
TABLE 4.2 Genetic distances among humans and the three great ape species, based on
sequence divergence in noncoding regions of DNA. (Data from M. A.
Jobling, M. Hurles, and C. Tyler-Smith, 2004, Human Evolutionary Genetics: Origins, Peoples &
Disease [New York: Garland Science], p. 214.)
A majority of biologists agree that as long as genetic distances are not too big or
too small, the molecular-clock assumption is a useful approximation. However, there
is some controversy about why genetic distance changes in a clocklike way. Advocates
of the neutral theory believe that most changes in DNA sequences have little or no
effect on fitness, so the evolution of this neutral DNA must be controlled by drift and
mutation. The neutral theory suggests that, under the right circumstances, mutation
and drift will produce clocklike change. Other biologists argue that the molecular
clock is the product of natural selection in variable environments.
The data from contemporary species indicate that genetic distance changes at a
constant rate, but the data don’t provide a clue as to what that rate might be. How-
ever, dated fossils allow us to estimate when the splits between lineages occurred. By
dividing the known genetic distance between a pair of species by the time since the last
common ancestor, we can estimate the rate at which genetic distance changes through
time. For example, fossil evidence (see Chapter 9) indicates that the last common
ancestor of orangutans and humans lived about 14 mya and that the genetic distance
between these two species is 3.1. Dividing the genetic distance between humans and
orangutans by the time since their last common ancestor indicates that genetic distance
accumulates at a rate of 0.22 unit per million years.
Once we have an estimate of the rate at which genetic distance changes through
time, the molecular-clock hypothesis can be used to date the divergence times for lin-
eages, even when we don’t have any fossils. For example, to estimate the last common
ancestor of humans and chimpanzees, we divide the genetic distance, 1.24, by the esti-
mated rate, 0.22. This calculation indicates that the last common ancestor of these two
species lived about 5.6 mya.
In practice, scientists use a number of different divergence dates to calibrate the
rate at which genetic distance changes through time. Each divergence date produces
a slightly different estimate of the rate of change of genetic distance, and this estimate
in turn generates different estimates of the divergence times. Thus, divergence dates for
humans and chimpanzees range from 7 to 5 mya.
Putting names on things is, to some extent, arbitrary. We could give species names like
Sam or Ruby, or perhaps use numbers like the Social Security system does. This is, in
fact, the way common names work. The word lion is an arbitrary label, as is
“Charles” or “550-72-9928.” The problem for scientists is that the number of animals
and plants is very large, encompassing far too many species for any individual to keep
track of. One way to cope with this massive complexity is to devise a system in which
organisms are grouped together in a hierarchical system of classification. Once again,
there are many possible systems. For example, we could categorize organisms alpha-
betically, grouping alligators and apricots with the As, barnacles and baboons with the
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 110
Bs, and so on. This approach has little to recommend it, because knowing how an ani-
mal is classified tells us nothing about the organism besides its location in the alpha-
bet. An alternative approach would be to adopt a system of classification that groups
together organisms with similar characteristics, analogous to the Library of Congress
system used by libraries to classify books. The Qs might be predators, the QHs aquatic
predators, the QPs aerial predators, and so on. Thus, knowing that the scientific name
of the red-tailed hawk is QP604.4 might tell you that the hawk is a small aerial pred-
ator that lives in North America. The problem with such a system is that not all organ-
isms would fall into a single category. Where would you classify animals that eat both
animals and plants, such as bears, or amphibious predators, such as frogs?
The scientific system for naming animals is based on the hierarchy of descent:
species that are closely related are classified together. Closely related species are
grouped together in the same genus (plural genera; Figure 4.30). For example, the
genus Pan contains two closely related species of chimpanzees: the common chim-
panzee, Pan troglodytes; and the bonobo, Pan paniscus. Closely related genera are
usually grouped together in a higher unit, often the family. Chimpanzees are in the
family Pongidae along with the other great apes: the orangutan (Pongo pygmaeus) and
the gorilla (Gorilla gorilla). Closely related families are then grouped together in a
more inclusive unit, often a superfamily. The Pongidae belong to the superfamily
Hominoidea along with the gibbons, siamangs, and humans.
The majority of taxonomists agree that descent should play a major role in clas-
sifying organisms. However, they vehemently disagree about whether descent should
be the only factor used to classify organisms. The members of a relatively new school
of thought, called cladistic taxonomy (or sometimes “cladistic systematics”), argue
that only descent should matter. Adherents to an older school of taxonomy, called evo-
lutionary taxonomy (or “evolutionary systematics”), believe that classification should
be based both on descent and on overall similarity. To understand the difference
between these two philosophies, consider Figure 4.31, in which humans have been
added to the phylogeny of the apes shown in Figure 4.30. Evolutionary taxonomists
would say that humans are qualitatively different from other apes and so deserve to
be distinguished at a higher taxonomic level (Figure 4.31a). Accordingly, these tax-
onomists classify humans in a family of their own, the Hominidae. For a cladist, this
is unacceptable because humans are descended from the same common ancestor as
Siamang Superfamily
Hominoidea
Gibbon
Orangutan Family
Bonobos and chimpanzees are Pongidae
FIGURE 4.30 Gorilla
classified together in the
genus Pan, both of these species are classified with Bonobo Genus
gorillas and orangutans in the family Pongidae, and all Pan
Chimpanzee
of the apes are grouped together in the superfamily
Hominoidea. Note that humans are missing from this
phylogenetic tree.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 111
Siamang Superfamily
Hominoidea
Gibbon
Orangutan
Family
Pongidae
Gorilla
Bonobo
Genus
Pan
Chimpanzee
Family
Human
Hominidae
Siamang Superfamily
Hominoidea
Gibbon
Orangutan Family
Pongidae
Gorilla
other members of the family Pongidae. This means that humans must be classified in
the same family as chimpanzees, bonobos, and gorillas (Figure 4.31b). It is not just
chauvinism about our own place in the primate phylogeny that causes discrepancies
between these classification schemes. The same problem arises in many other taxa. For
example, it turns out that crocodiles and birds share a more recent common ancestor
than either does with lizards. For a cladist, this means that birds and crocodiles must
be classified together, and lizards must be classified separately. Evolutionary taxono-
mists argue that birds are obviously distinctive and deserve a separate taxonomic
grouping.
In theory, cladistic taxonomy is both informative and unambiguous. It is inform-
ative because knowing an organism’s name and its position in the hierarchy of life tells
us how it is related to other organisms. It is unambiguous because the position of each
organism is given by the actual pattern of descent. Once you are confident that you
understand the phylogenetic relationships within a group, there is no doubt about how
any organism in that group should be classified. Cladists believe that evolutionary tax-
onomy is ambiguous because judgments of overall similarity are necessarily subjective.
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 112
On the other hand, evolutionary taxonomists complain that the advantages of cladis-
tics are mainly theoretical. In real life, they argue, uncertainty about phylogenetic rela-
tionships introduces far more ambiguity and instability in classification than do
judgments about overall similarity. For example, we are not completely certain
whether chimpanzees are more closely related to humans or to gorillas. Morpholog-
ical data suggest that chimpanzees are more closely related to gorillas. Although most
measures of genetic distance indicate that chimpanzees are more closely related to
humans, some genetic-distance data suggest the opposite. Given such uncertainty, how
would cladists name and classify these species?
It is important to keep in mind that this controversy is not about what the world
is like, or even about how evolution works. Instead it is a debate about how we should
name and classify organisms. Thus, no experiment or observation can prove either
school right or wrong. Instead, scientists must determine which system is more useful
in practice.
FURTHER READING
Dawkins, R. 1996. The Blind Watchmaker: Why the Evidence of Evolution Reveals
a Universe without Design. New York: Norton. Ch. 11.
Ridley, M. 1986. Evolution and Classification: The Reformation of Cladism. New
York: Longman.
———. 2004. Evolution. 3rd ed. Malden, Mass: Blackwell.
1. There are two large plants in the authors’ yard: a yucca tree and an enormous
prickly pear cactus. Both have rough, scaly bark, but it is known that these plants
are not closely related.
(a) What are two different explanations for the similarity of the bark?
(b) It is also known that yuccas are descended from a grass species instead of a
tree species. Which of the two explanations given in part 1a is consistent with
this fact?
2. Chimpanzees and gorillas more closely resemble each other anatomically than
either resembles humans. For example, the hands of chimpanzees and gorillas are
structurally similar and quite different from human hands. Genetic-distance data
suggest, however, that humans and chimpanzees are more closely related to each
other than either is to gorillas. Assuming that the genetic-distance data are cor-
rect, give two different explanations for the observed anatomical similarity
between chimpanzees and gorillas.
3. Use the genetic-distance matrix below to establish the taxonomic relationships
between the species listed. (Hint: Draw a phylogenetic tree to illustrate these tax-
onomic relationships.)
04_Speciation [3p]:How Humans Evolved [5e] 9/26/08 7:19 PM Page 113
A B C D
B 4.8 —
C 0.7 5.0 —
D 3.6 4.7 3.6 —
4. According to the biological species concept, what is a species? Why do some biol-
ogists define species in this way?
5. What is the ecological species concept? Why have some biologists questioned the
biological species concept?
6. Molecular methods allow biologists to measure the amount of gene flow among
populations that make up a species. When such methods first became available,
systematists were surprised to find that many morphologically indistinguishable
populations seem to be reproductively isolated from each other and thus, accord-
ing to the biological species concept, are considered entirely different species. Is
it possible to account for the existence of such cryptic species by allopatric speci-
ation? by parapatric speciation? by sympatric speciation?
7. New plant species are sometimes formed by the hybridization of existing species.
A new species retains all of the genes of each parent. For example, the variety of
wheat used to make bread is a hybrid of three different grass species. Explain how
such hybridization affects the family tree of these plants.
KEY TERMS
microevolution quadrupedal
macroevolution knuckle walking
species comparative method
testes terrestrial
biological species concept arboreal
reproductive isolation systematics
gene flow analogous
ecological species concept homologous
allopatric speciation ancestral trait
character displacement derived trait
reinforcement out-groups
parapatric speciation genetic distance
hybrid zone molecular clocks
sympatric speciation neutral theory
niche genus
adaptive radiation family
phylogeny superfamily
hominoids cladistic taxonomy
taxonomy evolutionary taxonomy
This page intentionally left blank
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 115
p a rt t w o
PRIMATE ECOLOGY
AND BEHAVIOR
Chapter 5
Two Reasons to Study Primates
Primates Are Our Closest Relatives
Primates Are a Diverse Order
Features That Define the Primates
Primate Biogeography
A Taxonomy of Living Primates
Primate Ecology
The Distribution of Food
Activity Patterns
Ranging Behavior
Predation
Primate Sociality
Primate Conservation
T
he chapters in Part Two focus on the behavior of living nonhuman primates.
Studies of nonhuman primates help us understand human evolution for two
complementary but distinct reasons. First, closely related species tend to be
similar morphologically. As we saw in Chapter 4, this similarity is due to the fact that
closely related species retain and share traits acquired through descent from a common
ancestor. For example, viviparity (bearing live young) and lactation are traits that all
placental and marsupial mammals share, and these traits distinguish mammals from
other taxa, such as reptiles. The existence of such similarities means that studies
116
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 117
of living primates often give us more insight about the behavior of our ancestors than
do studies of other organisms. This approach is called “reasoning by homology.” The
second reason we study primates is based on the idea that natural selection leads to
similar organisms in similar environments. By assessing the patterns of diversity in the
behavior and morphology of organisms in relation to their environments, we can see
how evolution shapes adaptation in response to different selective pressures. This
approach is called “reasoning by analogy.”
The fact that humans and other primates share many characteristics means that
other primates provide valuable insights about early humans.
We humans are more closely related to nonhuman primates than we are to any other
animal species. The anatomical similarities among monkeys, apes, and humans led the
Swedish naturalist Carolus Linnaeus to place us in the order Primates in the first sci-
entific taxonomy, Systema Naturae, published in 1735. Later, naturalists such as
Georges Cuvier and Johann Blumenbach placed us in our own order because of our
distinctive mental capacities and upright posture. In The Descent of Man, however,
Charles Darwin firmly advocated reinstating humans in the order Primates; he cited
biologist Thomas Henry Huxley’s essay enumerating the many anatomical similarities
between us and apes, and he suggested that “if man had not been his own classifier,
he would never have thought of founding a separate order for his own reception.”
Modern systematics unambiguously confirms that humans are more closely related to
other primates than to any other living creatures.
Because we are closely related to other primates, we share with them many aspects
of morphology, physiology, and development. For example, like other primates, we
have well-developed vision and grasping hands and feet. We share features of our life
history with other primates as well, including an extended period of juvenile develop-
ment and larger brains in relation to body size than the members of other taxonomic
groups. Homologies between humans and other primates also extend to behavior, since
the physiological and cognitive structures that underlie human behavior are more sim-
ilar to those of other primates than to members of other taxonomic groups. The exis-
tence of this extensive array of homologous traits, the product of the common
evolutionary history of the primates, means that nonhuman primates provide useful
models for understanding the evolutionary roots of human morphology and for unrav-
eling the origins of human nature.
Diversity within the primate order helps us to understand how natural selection
shapes behavior.
During the last 30 years, hundreds of researchers from a variety of academic disciplines
have spent thousands of hours observing many different species of nonhuman primates
in the wild, in captive colonies, and in laboratories. All primate species have evolved
adaptations that enable them to meet the basic challenges of life, such as finding food,
avoiding predators, obtaining mates, rearing young, and coping with competitors. At
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 118
the same time, there is great morphological, ecological, and behavioral diversity among
species within the primate order. For example, primates range in size from the tiny
mouse lemur, who weighs about 60 g (about 2 oz), to the male gorilla, who weighs about
260 times more—160 kg (350 lb). Some species live in dense tropical forests; others are
at home in open woodlands and savannas. Some subsist almost entirely on leaves; oth-
ers rely on an omnivorous diet that includes fruits, leaves, flowers, seeds, gum, nectar,
insects, and small animal prey. Some species are solitary, and others are highly gregar-
ious. Some are active at night (nocturnal); others are active during daylight hours (diur-
nal). One primate, the fat-tailed dwarf lemur, enters a torpid state and sleeps for six
months each year. Some species actively defend territories from incursions by other
members of their own species (conspecifics); others do not. In some species, only females
provide care of their young; in others, males participate actively in this process.
This variety is inherently interesting. Researchers who study primates tend to be
so absorbed in the lives of their subjects that they are motivated to endure the hard-
ships of fieldwork, the frustrations of attempting to obtain a share of ever-shrinking
research funds, and the puzzlement of family and friends who wonder why they have
chosen such an odd occupation. However, evidence of diversity among closely related
organisms living under somewhat different ecological and social conditions also
helps researchers to understand how evolution shapes behavior. Animals that are
closely related to one another phylogenetically tend to be very similar in morphology,
physiology, life history, and behavior. Thus, differences observed among closely
related species are likely to represent adaptive responses to specific ecological condi-
tions. At the same time, similarities among more distantly related creatures living
under similar ecological conditions are likely to be the product of convergence.
This approach, sometimes called the “comparative method,” has become an
important form of analysis as researchers attempt to explain the patterns of variation
in morphology and behavior observed in nature. The same principles have been bor-
rowed to reconstruct the behavior of extinct hominins, early members of the human
lineage. Because behavior leaves virtually no trace in the fossil record, the compara-
tive method provides one of our only objective means of testing hypotheses about the
lives of our hominin ancestors. For example, the observation that there are substan-
tial differences in male and female body size, a phenomenon called sexual dimorphism,
in species that form nonmonogamous groups suggests that highly dimorphic hominins
in the past were not monogamous. In Part Three, we will see how the data and theo-
ries about behavior produced by primatologists have played an important role in
reshaping our ideas about human origins.
The animals pictured in Figure 5.1 are all members of the primate order. These ani-
mals are similar in many ways: they are covered with a thick coat of hair, they have
four limbs, and they have five fingers on each hand. However, they share these ances-
tral features with all mammals. Beyond these ancestral features, it is hard to see what
the members of this group of animals have in common that makes them distinct from
other mammals. What distinguishes a ring-tailed lemur from a mongoose or a rac-
coon? What features link the langur and the aye-aye?
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 119
All of these
FIGURE 5.1
animals are pri-
mates: (a) aye-aye, (b) ring-tailed
lemur, (c) langur, (d) howler, (e)
gelada baboon. Primates are a diverse
order and do not possess a suite of
traits that unambiguously distinguish
them from other animals.
(d) (e)
In fact, primates are a rather nondescript mammalian order that cannot be unam-
biguously characterized by a single derived feature shared by all members. In his exten-
sive treatise on primate evolution, however, University of Chicago biologist Robert
Martin defines the primate order in terms of the derived features listed in Table 5.1.
The first three traits in Table 5.1 are related to the flexible movement of hands and
feet. Primates can grasp with their hands and feet (Figure 5.2a), and most monkeys and
apes can oppose their thumb and forefinger in a precision grip (Figure 5.2b). The flat
nails, distinct from the claws of many animals, and the tactile pads on the tips of pri-
mate fingers and toes further enhance their dexterity (Figure 5.2c). These traits enable
primates to use their hands and feet differently than most other animals do. Primates
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 120
TABLE 5.1 Definition of the primate order. See the text for more complete descriptions
of these features.
1. The big toe on the foot is opposable, and hands are prehensile. This means that primates can
use their feet and hands for grasping. The opposable big toe has been lost in humans.
2. There are flat nails on the hands and feet in most species, instead of claws, and there are
sensitive tactile pads with “fingerprints” on fingers and toes.
3. Locomotion is hind-limb–dominated, meaning that the hind limbs do most of the work, and the
center of gravity is nearer the hind limbs than the forelimbs.
4. There is an unspecialized olfactory (smelling) apparatus that is reduced in diurnal primates.
5. The visual sense is highly developed. The eyes are large and moved forward in the head,
providing stereoscopic vision.
6. Females have small litters, and gestation and juvenile periods are longer than in other mammals
of similar size.
7. The brain is large compared with the brains of similarly sized mammals, and it has a number of
unique anatomical features.
8. The molars are relatively unspecialized, and there is a maximum of two incisors, one canine,
three premolars, and three molars on each half of the upper and lower jaw.
9. There are a number of other subtle anatomical characteristics that are useful to systematists
but are hard to interpret functionally.
(b)
(a) Primates
FIGURE 5.2
have grasping
feet, which they use to climb, cling to
branches, hold food, and scratch them-
selves. (b) Primates can oppose the
thumb and forefinger in a precision grip
—a feature that enables them to hold
food in one hand while they are feeding,
to pick small ticks and bits of debris
from their hair while grooming, and (in
some species) to use tools. (c) Most
primates have flat nails on their hands
and sensitive tactile pads on the tips of
their fingers.
(a) (c)
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 121
PRIMATE BIOGEOGRAPHY
Primates are mainly restricted to tropical regions of the world.
The continents of Asia, Africa, and South America and the islands that lie near their
coasts are home to most of the world’s primates (Figure 5.5). A few species remain in
Mexico and Central America. Primates were once found in southern Europe, but no
natural population survives there now. There are no natural populations of primates
in Australia or Antarctica, and none occupied these continents in the past.
Primates are found mainly in tropical regions, where the fluctuations in temperature
from day to night greatly exceed fluctuations in temperature over the course of the year.
In the tropics, the distribution of resources that primates rely on for subsistence is affected
more strongly by seasonal changes in rainfall than by seasonal changes in temperature.
Some primate species extend their ranges into temperate areas of Africa and Asia, where
they manage to cope with substantial seasonal fluctuations in environmental conditions.
Within their ranges, primates occupy an extremely diverse set of habitats that
includes all types of tropical forests, savanna woodlands, mangrove swamps, grass-
lands, high-altitude plateaus, and deserts. The vast majority of primates, however, are
found in forested areas, where they travel, feed, socialize, and sleep in a largely arbo-
real world.
The distri-
FIGURE 5.5
bution of
living and fossil primates. Pri-
mates are now found in Central
America, South America, Africa,
and Asia. They are found mainly
in tropical regions of the world.
Primates were formerly found in
southern Europe and northern
Africa. There have never been
indigenous populations of pri-
Living Fossil only
mates in Australia or Antarctica.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 123
123
TABLE 5.2 A taxonomy of the living primates. (From R. Martin, 1992, “Classification of Primates,”
in S. Jones, R. Martin, and D. Pilbeam, eds., The Cambridge Encyclopedia of Human Evolution
[New York: Cambridge University Press], pp. 20–21.)
(b) (c)
(a) Marmosets are small-bodied New World monkeys who form monogamous
FIGURE 5.7
or polyandrous social groups. Males and older offspring participate actively in
the care of infants. Portraits of two cebid monkeys: (b) Spider monkeys rely heavily on ripe fruit and
travel in small parties. They have prehensile tails that they can use much like an extra hand or foot.
(c) Squirrel monkeys form large multimale, multifemale groups. In the mating season, males gain
weight and become “fatted,” and then compete actively for access to receptive females. (Pho-
tographs courtesy of Susan Perry (b) and Carlão Limeira (c).)
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 125
(a) Hanuman langurs are native to India and have been the subject of extensive
FIGURE 5.8
study during the last three decades. In some areas, hanuman langurs form one-
male, multifemale groups, and males engage in fierce fights over membership in bisexual groups. In these
groups, infanticide often follows when a new male takes over the group. Some representative cercop-
ithecines: (b) Bonnet macaques are one of several species of macaques that are found throughout Asia
and North Africa. Like other macaques, bonnet macaques form multimale, multifemale groups, and
females spend their entire lives in their natal (birth) groups. (c) Blue monkeys live in one-male, multife-
male groups. During the mating season, however, one or more unfamiliar males may join bisexual groups
and mate with females. (Photographs courtesy of Carola Borries (a), Kathy West (b), and Marina
Cords (c).)
of the earliest primates who lived 50 mya, they have many adaptations to living in
darkness, including a well-developed sense of smell, large eyes, and independently
movable ears (Figure 5.6). By contrast, monkeys and apes, who make up the subor-
der Anthropoidea, evolved adaptations more suited to a diurnal lifestyle early in their
evolutionary history. In the Anthropoidea, the traits related to increased complexity
of behavior are most fully developed. Anthropoid monkeys are generally larger than
prosimians, are active during the day, are more fully dependent on vision than smell,
and live in bigger and more complex social groups. The New World monkeys are all
arboreal, and some have prehensile tails (Figure 5.7). Old World monkeys typically live Gibbons live in
FIGURE 5.9
in one-male or multimale groups and span a wide range of ecological specializations, monogamous
activity patterns, and social organizations (Figure 5.8). The lesser apes, gibbons and groups and actively defend their territo-
siamangs, are slightly built and fully arboreal. They usually form pair-bonded groups ries against intruders. They have
and actively defend their territories (Figure 5.9). The great apes include bonobos and extremely long arms, which they use to
chimpanzees (Figure 5.10), our closest living relatives, as well as gorillas (Figure 5.11) propel themselves from one branch to
another as they swing hand over hand
and orangutans (Figure 5.12).
through the canopy, a form of locomo-
The classification of the primates into prosimians and anthropoids does not
tion called “brachiation.” Gibbons are
strictly reflect the patterns of descent among the animals in the suborders. Tarsiers are
confined to the tropical forests of Asia.
included in the prosimians because, like the lorises and lemurs, they are nocturnal crea- Like other residents of tropical forests,
tures that have retained many ancestral characters. However, both genetic and mor- their survival is threatened by the rapid
phological data suggest that tarsiers are more closely related to monkeys and apes than destruction of tropical forests. (Photo-
to prosimians. Thus, a purely cladistic classification would place tarsiers in the same graph courtesy of John Mitani.)
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 126
(a) (b)
infraorder (the taxonomic level immediately below suborder) as the monkeys. In fact,
many primate taxonomists advocate a taxonomy in which the lemurs and lorises are
classified together as strepsirhines and the rest of the primates are classified together
as haplorhines. The more traditional division into prosimians and anthropoids is an
example of evolutionary taxonomy in which both overall similarity and relatedness are
used to classify species.
PRIMATE ECOLOGY
Much of the day-to-day life of primates is driven by two concerns: getting enough to
eat and avoiding being eaten. Food is essential for growth, survival, and reproduction,
and it should not be surprising that primates spend much of every day finding, pro-
cessing, consuming, and digesting a wide variety of foods (Figure 5.13). At the same
time, primates must always guard against predators like lions, pythons, and eagles that
hunt them by day, and leopards that stalk them by night. As we will see in the chap-
ters that follow, both the distribution of food and the threat of predation influence the
extent of sociality among primates and shape the patterning of social interactions
within and between primate groups.
Below, we describe the basic features of primate ecology. Later we will draw on A female
this information to explore the relationships among ecological factors, social organi- FIGURE 5.13
baboon feeds
zation, and primate behavior. It is important to understand the nature of these rela- on corms in Amboseli, Kenya.
tionships because the same ecological factors are likely to have influenced the social
organization and behavior of our earliest ancestors.
Food provides energy that is essential for growth, survival, and reproduction.
Like all other animals, primates need energy to maintain normal metabolic processes;
to regulate essential body functions; and to sustain growth, development, and repro-
duction. The total amount of energy that an animal requires depends on four
components:
1. Basal metabolism. Basal metabolic rate is the rate at which an animal expends
energy to maintain life when at rest. As Figure 5.14 shows, large animals have
higher basal metabolic rates than small animals have. However, large animals
require relatively fewer calories per unit of body weight.
2. Active metabolism. When animals become active, their energy needs rise above
baseline levels. The number of additional calories required depends on how
Basal metabolic rate
much energy the animal expends. The amount of energy expended, in turn,
depends on the size of the animal and how fast it moves. In general, to sustain
a normal range of activities, an average-sized primate like a baboon or macaque
requires enough energy per day to maintain a rate about twice its basal metabolic
rate.
3. Growth rate. Growth imposes further energetic demands on organisms. Infants
and juveniles, who are gaining weight and growing in stature, require more
energy than would be expected on the basis of their body weight and activity lev-
els alone.
4. Reproductive effort. In addition, for female primates the energetic costs of
reproduction are substantial. During the latter stages of their pregnancies, for
example, primate females require about 25% more calories than usual; and dur-
ing lactation, about 50% more calories than usual.
A primate’s diet must satisfy the animal’s energy requirements, provide specific
types of nutrients, and minimize exposure to dangerous toxins.
The food that primates eat provides them with energy and essential nutrients, such
as amino acids and minerals, that they cannot synthesize themselves. Proteins are
essential for virtually every aspect of growth and reproduction and for the regulation
of many body functions. As we saw in Chapter 2, proteins are composed of long chains
of amino acids. Primates cannot synthesize amino acids from simpler molecules, so in
order to build many essential proteins, they must ingest foods that contain sufficient
amounts of a number of amino acids. Fats and oils are important sources of energy
for animals and provide about twice as much energy as equivalent volumes of carbo-
hydrates. Vitamins, minerals, and trace amounts of certain elements play an essential
role in regulating many of the body’s metabolic functions. Although they are needed
in only small amounts, deficiencies of specific vitamins, minerals, or trace elements can
cause significant impairment of normal body function. For example, trace amounts of
the elements iron and copper are important in the synthesis of hemoglobin, vitamin
D is essential for the metabolism of calcium and phosphorus, and sodium regulates the
quantity and distribution of body fluids. Primates cannot synthesize any of these com-
pounds and must acquire them from the foods they eat. Water is the major constituent
of the bodies of all animals and most plants. For survival, most animals must balance
their water intake with their water loss; moderate dehydration can be debilitating, and
significant dehydration can be fatal.
At the same time that primates attempt to obtain nourishment from food, they
must also take care to avoid toxins, substances in the environment that are harmful
to them. Many plants produce toxins called secondary compounds to protect them-
selves from being eaten. Thousands of these secondary compounds have been identi-
fied: caffeine and morphine are among the secondary compounds most familiar to us.
Some secondary compounds, such as alkaloids, are toxic to consumers because they
pass through the stomach into various types of cells, where they disrupt normal meta-
bolic functions. Common alkaloids include capsicum (the compound that brings tears
to your eyes when you eat red peppers) and chocolate. Other secondary compounds,
such as tannins (the bitter-tasting compound in tea), act in the consumer’s gut to reduce
the digestibility of plant material. Secondary compounds are particularly common
among tropical plant species and are often concentrated in mature leaves and seeds.
Young leaves, fruit, and flowers tend to have lower concentrations of secondary com-
pounds, making them relatively more palatable to primates.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 129
TABLE 5.3 Sources of nutrients for primates. (x) indicates that the nutrient content is
generally accessible only to animals that have specific digestive adaptations.
Animals x (x) x x x x
Fruit x x
Seeds x x x
Flowers x x
Young leaves x x x x
Mature leaves (x)
Woody stems x
Sap x x x
Gum x (x) x
Underground parts x x x
Primates obtain energy and essential nutrients from a variety of sources (Table
5.3). Carbohydrates are obtained mainly from the simple sugars in fruit, but animal
prey, such as insects, also provide a good source of fats and oils. Gum, a substance that
plants produce in response to physical injury, is an important source of carbohydrates
for some primates, particularly galagos, marmosets, and tamarins. Primates get most
of their protein from insect prey or from young leaves. Some species have special adap-
tations that facilitate the breakdown of cellulose, enabling them to digest more of the
protein contained in the cells of mature leaves. Although seeds provide a good source
of vitamins, fats, and oils, many plants package their seeds in husks or pods that shield
their contents from seed predators. Many primates drink daily from streams, water
holes, springs, or puddles of rainwater (Figure 5.15). Primates can also obtain water
from fruit, flowers, young leaves, animal prey, and the underground storage parts
These savanna
FIGURE 5.15
baboons are
drinking from a pool of rainwater. Most
primates must drink every day.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 130
(roots and tubers) of various plants. These sources of water are particularly important
for arboreal animals that do not descend from the canopy of tree branches and for ter-
restrial animals during times of the year when surface water is scarce. Vitamins, min-
erals, and trace elements are obtained in small quantities from many different sources.
Although primates display considerable diversity in their diet, some generaliza-
tions are possible:
1. All primates rely on at least one type of food that is high in protein and another
that is high in carbohydrates. Prosimians generally obtain protein from insects
and carbohydrates from gum and fruit. Monkeys and apes usually obtain pro-
tein from insects or young leaves and carbohydrates from fruit.
2. Most primates rely more heavily on some types of foods than on others. Chim-
panzees, for example, feed mainly on ripe fruit throughout their range from Tan-
zania to the Ivory Coast. Scientists use the terms frugivore, folivore, insectivore,
and gummivore to refer to primates who rely most heavily on fruit, leaves,
insects, and plant gum, respectively. Box 5.1 examines some of the adaptations
in morphology among primates with different diets.
3. In general, insectivores are smaller than frugivores, and frugivores are smaller than
folivores (Figure 5.16). These differences in size are related to the fact that small
animals have relatively higher energy requirements than larger animals do, and
4
Number of species
3
2
1
1 kg 10 kg 100 kg
Insects and gum
4
Number of species
3
2
1
1 kg 10 kg 100 kg
Fruit and some insects
6
Number of species
5
4
3
2
1
1 kg 10 kg 100 kg
Fruit and some leaves
6
Number of species
5
4
3
2
1
1 kg 10 kg 100 kg
Leaves, seeds, and herbs
Box 5.1
Teeth And Guts: You Are
What You Can Chew
For various reasons, biological anthropologists spend a lot called the dental formula, which is commonly written in the
of time thinking about teeth. Teeth are useful markers for following form:
taxonomic identity, since various kinds of primates have dif- 2.1.3.3
ferent numbers of teeth. Teeth are also useful because they 2.1.3.3
tell us things about what kinds of food primates eat. If we Reading from left to right, the numerals tell us how many
can detect a relationship between dental morphology and incisors, canines, premolars, and molars a particular species
diet, we can apply these insights to the fossil record. This is has (or had) on one side of its jaw. The top line of numbers
particularly handy because teeth are the most commonly represents the teeth on one side of the upper jaw (maxilla),
preserved parts of the body. Finally, teeth and gut mor- and the bottom line represents the teeth on the correspon-
phology provide examples of how natural selection has ding side of the lower jaw (mandible). Usually, but not
created adaptations that enable animals to cope with their always, the formula is the same for both upper and lower
environments more effectively. jaws. Like most other parts of the body, our dentition is
bilaterally symmetrical, which means that the left side is
identical to the right side. The ancestral pattern shown here
has been modified in various primate taxa, as the total num-
DENTAL FORMULA ber of teeth has been reduced.
The dental formulas among living primates vary (Table
To appreciate the basic features of primate dentition, you 5.4). The lorises, pottos, galagos, and a number of lemurids
can consult Figure 5.17, or you can simply look in a mirror, have retained the primitive dental formula, but other
since your teeth are much like those of other primates. Teeth prosimian taxa have lost incisors, canines, or premolars.
are rooted in the jaw. The jaw holds four different kinds of Tarsiers have lost one incisor on the mandible but have
teeth: in order, they are, first, the incisors at the front; then retained two on the maxilla. All of the New World mon-
come the canines, premolars, and the molars in the rear. All keys, except the marmosets and tamarins, have retained the
primates have the same kinds of teeth, but species vary in primitive dental formula; the marmosets and tamarins have
how many of each kind of tooth they have. For convenience, lost one molar. The Old World monkeys, apes, and humans
these combinations are expressed in a standard format have only two premolars.
Incisors
Incisors
Canine
Canine
Cusps
131
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 132
they require relatively small amounts of high-quality foods that can be processed
quickly. Larger animals are less constrained by the quality of their food than by
the quantity because they can afford to process lower-quality foods more slowly.
Cerocebus
Long
small intestine
Lepilemur Colobus
Cercocebus
Fruit eaters
Phaner Enlarged
large intesine
Callithrix
Macaca
Stout
incisors Leaf eaters
Callimico
Sharp cusps
replenished. Instead, the availability of their preferred foods varies widely in space and
time, making their food sources patchy and often unpredictable. Most primate
species live in tropical forests. Although such forests, with their dense greenery, seem
to provide abundant supplies of food for primates, appearances can be deceiving.
Tropical forests contain a very large number of tree species, and individual trees of
any particular species are few in number.
Primates with different dietary specializations confront different foraging chal-
lenges (Figure 5.19). Plants generally produce more leaves than flowers or fruit, and
they bear leaves for a longer period during the year than they bear flowers and fruit.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 134
(a) (b)
(c) (d)
(e) (f)
(a) Some primates feed mainly on leaves, though many leaves contain toxic
FIGURE 5.19
secondary plant compounds. The monkeys shown here are red colobus monkeys
in the Kibale Forest of Uganda. (b) Some primates include a variety of insects and other animal prey in
their diet. This capuchin monkey in Costa Rica is feeding on a wasps’ nest. (c) Mountain gorillas are
mainly vegetarians. They consume vast quantities of plant material, like this fibrous stem. (d) This
vervet monkey is feeding on grass stems. (e) Although many primates feed mainly on one type of food,
such as leaves or fruit, no primate relies exclusively on one type of food. For example, the main bulk of
the muriqui diet comes from fruit, but muriquis also eat leaves, as shown here. (f) Langurs are folivores.
Here, hanuman langurs in Ramnagar, Nepal, forage for water plants. (Photographs courtesy of Lynne
Isbell (a), Susan Perry (b), John Mitani (c), Carlão Limeira (e), and Carola Borries (f).)
As a result, foliage is normally more abundant than fruit or flowers are at a given time
during the year, and mature leaves are more abundant than young leaves. Insects and
other suitable prey animals occur at even lower densities than plants. This means that
folivores can generally find more food in a given area than frugivores or insectivores
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 135
can. However, the high concentration of toxic secondary compounds in mature leaves
complicates the foraging strategies of folivores. Some leaves must be avoided alto-
gether, and others can be eaten only in small quantities. Nonetheless, the food
supplies of folivorous species are generally more uniform and predictable in space and
time than are the food supplies of frugivores or insectivores. Thus, it is not surprising
to find that folivores generally have smaller home ranges than frugivores or insecti-
vores have.
Activity Patterns
Primates spend the majority of their time feeding, moving around their home ranges,
and resting (Figures 5.20 and 5.21a). Relatively small portions of each day are spent
grooming, playing, fighting, or mating (Figure 5.21b). The proportion of time devoted
to various activities is influenced to some extent by ecological conditions. For primates
living in seasonal habitats, for example, the dry season is often a time of scarce
resources, and it is harder to find adequate amounts of appropriate types of food. In
some cases, this means that the proportion of time spent feeding and traveling
increases during the dry season, while the proportion of time spent resting and social-
izing decreases.
Primate activity also shows regular patterns over the course of the day. When pri-
mates wake up, their stomachs are empty, so the first order of the day is to visit a feed-
ing site. Much of the morning is spent eating and moving between feeding sites. As the
sun moves directly overhead and the temperature rises, most species settle down in a
shady spot to rest, socialize, and digest their morning meals (Figure 5.21c). Later in
the afternoon they resume feeding. Before dusk they move to the night’s sleeping site;
some species sleep in the same trees every night; others have multiple sleeping sites
within their ranges.
Brown
capuchin
White-fronted
capuchin
Squirrel
monkey
Emperor
tamarin
Saddleback
tamarin
(a) (b)
(c)
Ranging Behavior
All primates have home ranges, but only some species are territorial—defending
their home range against incursions by other members of their species.
In all primate species, groups range over a relatively fixed area, and members of a given
group can be consistently found in a particular area over time. These areas are called
home ranges, and they contain all of the resources that group members exploit in feed-
ing, resting, and sleeping. However, the extent of overlap among adjacent home ranges
and the nature of interactions with members of neighboring groups or strangers vary
considerably among species. Some primate species, like gibbons, maintain exclusive
access to fixed areas, called territories (Figure 5.22). Territory residents regularly adver-
tise their presence by vocalizing, and they aggressively protect the boundaries of their
territories from encroachment by outsiders (Figure 5.23). Although some territorial
birds defend only their nest sites, primate territories contain all of the sites at which
the residents feed, rest, and sleep and the areas in which they travel. Thus, among ter-
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 137
ritorial primates, the boundaries for the territory are essentially the same as for their
home range, and territories do not overlap.
Nonterritorial species, like squirrel monkeys and long-tailed macaques, establish
home ranges that overlap considerably with those of neighboring groups (Figure 5.22).
When members of neighboring nonterritorial groups meet, they may fight (Figure
5.24), exclude members of lower-ranking groups from resources, avoid one another,
or mingle peacefully together. This last option is unusual, but in some species, adult
females sexually solicit males from other groups, males attempt to mate with females
from other groups, and juveniles from neighboring groups play together when their
groups are in proximity.
The two main functions suggested for territoriality are resource defense and
mate defense.
To understand why some primate species defend their home ranges from intrud-
ers and others do not, we need to think about the possible costs and benefits associ-
ated with defending resources from conspecifics. Costs and benefits are measured in
terms of the impact on the individual’s ability to survive and reproduce successfully.
Territoriality is beneficial because it prevents outsiders from exploiting the limited
resources within a territory. At the same time, however, territoriality is costly because
the residents must be constantly vigilant against intruders,
regularly advertise their presence, and be prepared to defend
their ranges against encroachment. Territoriality is expected
to occur only when the benefits of maintaining exclusive
access to a particular piece of land outweigh the costs of pro-
tecting these benefits.
When will the benefits of territoriality exceed the costs?
The answer to this question depends in part on the kinds of
resources that individuals need in order to survive and repro-
duce successfully, and in part on the way these resources are
distributed spatially and seasonally. For reasons we will dis-
cuss more fully in Chapter 6, the reproductive strategies of
mammalian males and females generally differ. In most
cases, female reproductive success depends mainly on getting
enough to eat for themselves and their dependent offspring, Gibbons perform complex vocal duets as part
FIGURE 5.23
and males’ reproductive success depends mainly on their of territorial defense.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 138
ability to mate with females. As a consequence, females are more concerned about
access to food, and males are more interested in access to females. Thus, territoriality
has two different functions. Sometimes females defend food resources, or males
defend food resources on their behalf. Other times, males defend groups of females
against incursions by other males. In primates, both resource defense and mate
defense seem to have influenced the evolution of territoriality.
Predation
Predation is believed to be a significant source of mortality among primates,
but direct evidence of predation is difficult to obtain.
Primates are hunted by a wide range of predators, including pythons, raptors, croco-
diles, leopards, lions, tigers, and humans (Figure 5.25). In Madagascar, large lemurs
are preyed upon by fossas, pumalike carnivores. Primates are also preyed on by other
primates. Chimpanzees, for example, hunt red colobus monkeys, and baboons prey
on vervet monkeys.
The estimated rates of predation vary from less than 1% of the population per
year to more than 15%. The available data suggest that small-bodied primates are
more vulnerable to predation than larger ones are and that immature primates are
generally more susceptible to predation than adults are. These data are not very
solid, however, because systematic information about predation is quite hard to
come by since most predators avoid close contact with humans, and some predators,
like leopards, generally hunt at night, when most researchers are asleep. Usually pre-
dation is inferred when a healthy animal that is unlikely to have left the group
abruptly vanishes without a trace (Figure 5.26). Such inferences are, of course, sub-
ject to error.
Another approach is to study the predators, not their prey. Crowned hawk eagles
(Figure 5.25d) are the only large raptors that live in the tropical rain forests of Africa.
They are formidable predators; although they weigh only 3 to 4 kg (6.6 to 8.8 lb), they
have powerful legs and large talons, and can take prey that weigh up to 20 kg (44 lb).
Crowned hawk eagles carry prey back to their nests and discard the bones. By sort-
ing through the remains under crowned hawk eagle nests, researchers can figure out
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 139
(e)
(d)
what they eat. Analyses of nest remains in the Kibale Forest of Uganda and the Taï For-
est in Ivory Coast indicate that crowned hawk eagles prey on all of the primates in
these forests except chimpanzees. Monkeys make up 60% to 80% of the crowned
hawk eagles’ diets at these sites, and the eagles kill a sizable fraction (2% to 16%) of
the total populations of various primate species in these forests each year.
Susanne Schultz of the University of Liverpool and her colleagues have compared
the characteristics of mammalian prey taken by crowned hawk eagles, leopards,
and chimpanees in the Taï Forest (Figure 5.27). In general, terrestrial species are more
vulnerable than arboreal species, and species that live in small groups are more
In some cases, researchers are able to confirm predation. Here, an adult female
FIGURE 5.26
baboon in the Okavango Delta, Botswana, was killed by a leopard. You can see
(a) the depression in the sand that was made when the leopard dragged the female’s body out of the
sleeping tree and across a small sandy clearing, (b) the leopard’s footprints beside the drag marks, and
(c) the remains of the female the following morning—her jaw, bits of her skull, and clumps of hair.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 140
0.25
0.20
Predation rate
0.15
0.10
0.5
0
us
us
on
us
tto
y
e
ke
be
nz
ob
ob
ob
en
Po
ga
on
pa
ol
l
gu
co
co
an
am
ec
im
ed
d
te
m
iv
Re
Ch
an
hi
os
ed
Ol
w
Di
-n
cl
d
te
ta
an
hi
ec
w
Sp
ac
er
bl
ss
Le
rn
te
es
W
The rate of predation by leopards (red), eagles (tan), and chimpanzees (blue)
FIGURE 5.27
in the Taï Forest on different primate species is shown here. Note that the
preferred prey of chimpanzees are red colobus monkeys, and chimpanzees’ only predators are leopards.
(From S. Schultz, R. Noë, W. S. MacGraw, and R. I. M. Dunbar, 2004, “A Community-Level Evalua-
tion of the Impact of Prey Behavioural and Ecological Characteristics on Predator Diet Composition,”
Proceedings of the Royal Society of London, Section B 271:725–732.)
vulnerable than animals that live in large groups. Thus, arboreal monkeys that live in
large groups face the lowest risks. Schultz and her colleagues suggest that these results
may explain some aspects of the distribution of terrestrial primates in Africa, Asia, and
the neotropics. In Africa, with crowned hawk eagles and at least two large predatory
felids, terrestial primates are large-bodied or live in large groups. In Asia, where there
are no large forest raptors and few large felids, there are several semiterrestrial
macaque species. And in the neotropics, where there are several species of large felids
and forest raptors, there are no terrestrial monkeys at all.
Many primates give alarm calls when they sight potential predators, and some
species have specific vocalizations for particular predators. Vervet monkeys, for
example, give different calls when they are alerted to the presence of leopards, small
carnivores, eagles, snakes, baboons, and unfamiliar humans. In many species, the most
common response to predators is to flee or take cover. Small primates sometimes try
to conceal themselves from predators; larger ones may confront potential predators.
When slow-moving pottos encounter snakes, for example, they fall to the ground,
move a short distance, and freeze. At some sites, adult red colobus monkeys aggres-
sively attack chimpanzees who hunt their infants.
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 141
PRIMATE SOCIALITY
Sociality has evolved in primates in response to ecological pressures. Social life
has both costs and benefits.
Nearly all primates live in social groups of one kind or another. Sociality has evolved
in primates because there are important benefits associated with living in groups. Pri-
mates that live in groups may be better able to acquire and control resources. Animals
that live in groups can chase lone individuals away from feeding trees and can protect
their own access to food and other resources against smaller numbers of intruders. As
we saw above, grouping also provides safety from predators. This is because of the
three Ds: detection, deterrence, and dilution. Animals in groups are more likely to
detect predators because there are more pairs of eyes on the lookout for predators.
Animals in groups are also more effective in deterring predators by actively mobbing
or chasing them away. Finally, the threat of predation to any single individual is diluted
when predators strike at random. If there are two animals in a group, and a predator
strikes, each animal has a 50% chance of being eaten. If there are 10 individuals, the
individual risk is decreased to 10%.
Although there are important benefits associated with sociality, there are equally
important costs. Animals that live in groups may encounter more competition over
access to food and mates, become more vulnerable to disease, and face various haz-
ards from conspecifics (such as cannibalism, cuckoldry, incest, or infanticide).
The size and composition of the groups that we see in nature are expected to
reflect a compromise between the costs and benefits of social-
ity for individuals. The magnitude of these costs and bene-
fits is influenced by both social and ecological factors.
Box 5.2
Forms of Social Groups
among Primates
Although most primates are social, their groups vary in size Polyandry. One female is paired with two or more males.
and composition. One way to order this diversity is to clas- They share a territory or home range with their offspring.
sify primate social systems according to the mating and res- Polyandry may occur among some of the marmosets and
idence patterns of females and males. The major forms of tamarins.
primate social systems are described here and diagrammed
Polygyny: one-male. Groups are composed of several
in Figure 5.29:
adult females, a single resident male, and immature indi-
Solitary. Females maintain separate home ranges or ter- viduals. In species with this form of polygyny, males who
ritories and associate mainly with their dependent off- do not reside in bisexual groups often form all-male
spring. Males may establish their own territories or home groups. These all-male groups may mount vigorous
ranges, or they may defend the ranges of several adult assaults on resident males of bisexual groups in an
females from incursions by other males. With the excep- attempt to oust residents. Dispersal patterns are variable
tion of orangutans, all of the solitary primates are in species that form one-male, multifemale groups. One-
prosimians. male, multifemale groups are common among howlers,
langurs, and gelada baboons.
Pair-bonded. One adult male and one female form a
mating relationship and share a territory with their Polygyny: multimale. Groups are composed of several
immature offspring. Pair bonding is characteristic of gib- adult males, adult females, and immature animals.
bons, some of the small New World monkeys, and a few In most species that form multimale, multifemale groups,
prosimian species. males disperse from their natal groups while females
Solitary Polygyny:
one-male
The major
FIGURE 5.29
types of social
groups that primates form. When males
and females share their home ranges,
Pair-bonded
their home ranges are drawn here in
lavender. When the ranges of the two
sexes differ, male home ranges are
Polygyny:
drawn in blue and female home ranges multimale
are drawn in magenta. The sizes of the
male and female symbols reflect the
degree of sexual dimorphism among
males and females. Polyandry
142
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 143
remain. Dispersing males may move directly between partners sometimes copulate with nonresidents; and in
bisexual groups, remain solitary for short periods of species that normally form multimale groups, some groups
time, or join all-male groups. Multimale, multifemale may have only one adult male. Moreover, some species do
groups are characteristic of macaques, baboons, vervets, not fit neatly within this classification system. Female chim-
squirrel monkeys, capuchin monkeys, and some panzees and spider monkeys occupy individual home ranges
colobines. and spend much of their time alone or with their dependent
offspring. In these species, multiple males jointly defend the
This taxonomy of social groups provides one way to home ranges of multiple females. We call these fission–
classify primate social organization, but readers should rec- fusion groups because party size is variable. Hamadryas and
ognize that these categories are idealized descriptions of res- gelada baboons form one-male, multifemale units, but sev-
idence and mating patterns. The reality is inevitably more eral one-male, multifemale units collectively belong to larger
complicated. For example, in some pair-bonded groups, aggregations.
groups, and this may outweigh any advantages gained in competition between groups.
Thus, primates might live in groups because groups provide greater safety from
predators.
The data are inconclusive and so the jury is still out on whether resource compe-
tition or predation was the primary factor favoring the evolution of sociality in pri-
mates. However, many primatologists are convinced that the nature of resource
competition affects the behavioral strategies of primates, particularly females, and
influences the composition of primate groups (Box 5.2). Females come first in this sce-
nario because their fitness depends mainly on their nutritional status: well-nourished
females grow faster, mature earlier, and have higher fertility rates than do poorly nour-
ished females. In contrast, males’ fitness depends primarily on their ability to obtain
access to fertile females, not on their nutritional status. Thus, ecological pressures influ-
ence the distribution of females, and males distribute themselves to maximize their
access to females. (We will discuss male and female reproductive strategies more fully
in Chapter 6.)
PRIMATE CONSERVATION
Many species of primates are endangered by (1) habitat destruction,
(2) hunting, or (3) live capture for trade and export.
Tropical deforestation
Tropical forest
This map shows the major locations of tropical forest and the areas that have
FIGURE 5.31
become deforested (90% of the canopy cover has been lost). Deforestation is
a major threat to primates because many primate species live in tropical forests.
As arboreal residents of the tropics, most primate populations are directly affected
by the rapid and widespread destruction of the world’s forests. Primate ecologists
Colin Chapman of the University of Florida and Carlos Peres of the University of East
Anglia recently reviewed the conservation status of the world’s primate populations.
Their analysis is quite sobering. Between 1980 and 1995, approximately 10% of the
forests in Africa and Latin America were lost, and 6% of the forests of Asia disap-
peared (Figure 5.31). Countries that house primates are losing about 125,000 km2 of
forest—an area about the size of the state of Mississippi—every year.
The destruction of tropical forests is the product of economic and demographic
pressures acting on governments and local residents. Many developing countries have
huge foreign debts that must be repaid. The need to raise funds to pay off these debts
generates intense pressure for timber harvesting and more intensive agricultural activ-
ity. Each year, 5 million to 6 million hectares of forest are logged, seriously disrupting
the lives of the animals that live in them. (A hectare is a square measuring 100 m on
a side, or about 2.5 acres.)
Forests are also cleared for agricultural activities. Rapid increases in the popula-
tion of underdeveloped countries in the tropics have created intense demand for addi-
tional agricultural land. In West Africa, Asia, and South America, for example, vast
expanses of forests have been cleared to accommodate the demands of subsistence farm-
ers trying to feed their families, as well as the needs of large-scale agricultural projects.
In Central and South America, massive areas have been cleared for large cattle ranches.
In the last two decades, a new threat to the forests of the world has emerged: wild-
fire. Major fires have destroyed massive tracts of forest in Southeast Asia and South
05_Intro to Primates [4p]:How Humans Evolved [5e] 10/10/08 11:36 AM Page 145
America. Ecologists believe that natural fires in tropical forests are relatively rare, and
that these devastating fires are the product of human activity. In Indonesia, massive
fires in the late 1990s left thousands of orangutans dead, reducing their numbers by
nearly a third.
In many areas around the world, particularly South America and Africa, primates
are also hunted for meat. Although systematic information about the impact of hunt-
ing on wild primate populations is scant, some case studies reveal troubling trends. In
one forest in Kenya, for example, 1,200 blue monkeys and nearly 700 baboons were
killed by subsistence hunters in one year. In the Brazilian Amazon, one family of rub-
ber tappers killed 200 woolly monkeys, 100 spider monkeys, and 80 howler monkeys
during an 18-month span. In addition to subsistence hunting, there is also an active
market for “bushmeat” in many urban areas.
The capture and trade of live primates has been greatly reduced since the Con-
vention on International Trade in Endangered Species of Wild Fauna and Flora
(CITES) was drafted in 1973. The parties to CITES, which now number 167 countries,
ban commercial trade of all endangered species and monitor the trade of those that
are at risk of becoming endangered. CITES has been an effective weapon in protect-
ing primate populations around the world. The United States imported more than
100,000 primates each year before ratifying CITES but had reduced this number to
approximately 13,000 a decade after signing the international agreement.
Although CITES has made a major impact, some problems persist. Live capture
for trade remains a major threat to certain species, particularly the great apes, whose
high commercial value creates strong incentives for illegal commerce. In many com-
munities, young primates are kept as pets (Figure 5.32). For each animal taken into
captivity, many other animals are put at risk because hunters cannot obtain young
primates without capturing their mothers, who are usually killed in the process. In
addition, many prospective pets die after capture from injuries suffered during cap-
ture and transport or from poor housing conditions and inappropriate diets while in
captivity.
Efforts to save endangered primate populations have met with some success.
FURTHER READING
Boinski, S., and P. A. Garber, eds. 2000. On the Move: How and Why Animals Travel
in Groups. Chicago: University of Chicago Press.
Campell, C. J., A. Fuentes, K. C. MacKinnon, M. Panger, and S. K. Bearder, eds. 2007.
Primates in Perspective. New York: Oxford University Press.
Cowlishaw, G., and R. I. M. Dunbar. 2000. Primate Conservation Biology. Chicago:
University of Chicago Press.
Hohmann, G., M. M. Robbins, and C. Boesch, eds. 2006. Feeding Ecology in Apes
and Other Primates: Ecological, Physical, and Behavioral Aspects. New York:
Cambridge University Press.
Kappeler, P. M., and M. Pereira, eds. 2003. Primate Life Histories and Socioecology.
Chicago: University of Chicago Press.
Kramer, R., C. van Schaik, and J. Johnson, eds. 1997. Last Stand: Protected Areas and
the Defense of Tropical Biodiversity. New York: Oxford University Press.
Lee, P. C., ed. 1999. Comparative Primate Socioecology. New York: Cambridge Uni-
versity Press.
Martin, R. D. 1990. Primate Origins and Evolution: A Phylogenetic Reconstruction.
Princeton: Princeton University Press.
Strier, K. B. 2007. Primate Behavioral Ecology. 3rd ed. Boston: Allyn & Bacon.
10. Some primatologists have predicted that there should be a linear effect of group
size on vigilance rates, if sociality evolved as a defense against predation. This
means that every increase in the number of animals in the group should be asso-
ciated with a decrease in the proportion of time that animals spend in vigilance
activities. Many studies have failed to show a linear effect. How would you inter-
pret these results? What do the results tell you about the selective forces that shape
sociality in primates?
11. Sociality is a relatively uncommon feature in nature. What are the potential advan-
tages and disadvantages of living in social groups? Why are (virtually all) primates
social?
12. The future of primates, and other occupants of tropical forests, is precarious.
What are the major hazards that primates face?
13. How can we balance the needs and rights of people living in the developing
nations of the tropics with the needs of the animals who live in tropical forests?
KEY TERMS
Chapter 6
The Language of Adaptive Explanations
The Evolution of Reproductive Strategies
Reproductive Strategies of Females
Sources of Variation in Female
Reproductive Performance
Reproductive Trade-offs
Sexual Selection and Male Mating Strategies
Intrasexual Selection
Male Reproductive Tactics
Investing Males
Male–Male Competition in Groups
without Pair Bonds
Infanticide
R
eproduction is the central act in the life of every living thing. Primates per-
form a dizzying variety of behaviors: gibbons fill the forest with their haunt-
ing duets, baboons threaten and posture in their struggle for dominance over
other members of their group, and chimpanzees use carefully selected stone hammers
to crack open tough nuts. But all of these behaviors evolved for a single ultimate pur-
pose: reproduction. According to Darwin’s theory, complex adaptations exist because
they evolved step by step by natural selection. At each step, only those modifications
that increased reproductive success were favored and retained in subsequent genera-
tions of offspring. Thus, every morphological feature and every behavior exist
148
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 149
Biologists often use the term strategy to describe the behavior of animals. For exam-
ple, folivory is characterized as a foraging strategy, and monogamy is described as a
mating strategy. When evolutionary biologists use the term, they mean something very
different from what we normally mean when we use strategy to describe, say, a gen-
eral’s military maneuvers or a baseball manager’s tactics. In common usage, strategy
implies a conscious plan of action. Evolutionary biologists do not think other animals
consciously decide to defend their territories, wean their offspring at a particular age,
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 150
monitor their ingestion of secondary plant compounds, and so on. Instead, strategy
refers to a set of behaviors occurring in a specific functional context, such as mating,
parenting, or foraging. Strategies are the product of natural selection acting on indi-
viduals to shape the motivations, reactions, preferences, capacities, and choices that
influence behavior. Strategies that led to greater reproductive success in ancestral pop-
ulations have been favored by natural selection and represent adaptations.
The amount of parental care varies greatly within the animal kingdom. In most
species, parents do little for their offspring. For example, most frogs lay their eggs and
never see their offspring again. In such species, the nutrients that females leave in the
egg are the only form of parental care. In contrast, primates—like almost all birds
and mammals, and like some invertebrates and fish—provide much more than just
the resources included in gametes. At least one parent—and sometimes both—
shelters its young from the elements, protects them from predators, and provides
them with food.
The relative amount of parental care provided by mothers and fathers also varies
within the animal kingdom. In species without parental care, females produce large,
nutrient-rich gametes, and males produce small gametes and supply only genes.
Among species with parental care, however, all possible arrangements occur. Primate
mothers always nurse their offspring and often provide extensive care (Figure 6.1). The
In all primate behavior of fathers is much more variable. In most species, fathers give nothing to their
FIGURE 6.1 offspring other than the genes contained in their sperm. In a minority of species, how-
species, females
nurse their young. In baboons and many ever, males are devoted fathers. In other taxa, patterns differ. For example, in most bird
other species, females provide most of species, males and females form pairs and raise their young together (Figure 6.2), and
the direct care that infants receive. there are even some bird species in which only males care for chicks.
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 151
Males do not care for their offspring (1) when they can
easily use their resources to acquire many additional In most species of birds, the male and female
FIGURE 6.2
matings or (2) when caring for their offspring would form a pair bond and jointly raise their young.
not appreciably increase the offspring’s fitness. Here a bald eagle carries food to its hungry brood.
At first glance, it seems odd that most primate males fail to provide any care for
their offspring. Surely, if the males helped their mates, they would increase the
chances that their offspring would survive to adulthood. Therefore, we might expect
paternal care to be favored by natural selection (Figure 6.3).
If time, energy, and other resources were unlimited, this reasoning would be cor-
rect. In real life, though, time, energy, and material resources are always in short sup-
ply. The effort that an individual devotes to caring for offspring is effort that cannot
be spent competing for prospective mates. Natural selection will favor individuals that
allocate effort among these competing demands so as to maximize the number of sur-
viving offspring that they produce.
To understand the evolution of unequal parental investment, we must identify the
conditions under which one sex can profitably reduce its parental effort at the expense
of its partner. Consider a species in which most males help their mates feed and care
for their offspring. Even in such a species, a few males will have a heritable tendency
to invest less in their offspring. We will refer to these two types as “investing” and
“noninvesting” fathers. Because time, energy, and resources are always limited, males
that devote more effort to caring for offspring must allocate less effort to competing
for access to females. On average, the offspring of noninvesting males will receive less
care than will the offspring of investing males, making them less likely to survive and
to reproduce successfully when they mature. On the other hand, because noninvest-
ing males are not kept busy caring for their offspring, on average they will acquire
more mates than will investing males. Mutations favoring the tendency to provide less
parental care will increase in frequency when the benefits to males (measured in terms
of the increase in fitness gained from additional matings) outweigh the costs to males
(measured as the decrease in offspring fitness due to a reduction in paternal care). In most non-
FIGURE 6.3
This reasoning suggests that unequal parental investment will be favored when monogamous
one or both of the following are true: species, males have relatively little con-
tact with infants. Although males like
1. Acquiring additional mates is relatively easy, so considerable gains are achieved this bonnet macaque are sometimes
by allocating additional effort to attracting mates. quite tolerant of infants, they rarely
2. The fitness of offspring raised by only one parent is high, so the payoff for addi- carry, groom, feed, or play with infants.
tional parental investment is relatively low. (Photograph by Kathy West.)
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 152
The key factors are the costs of finding additional mates and the benefits associ-
ated with incremental increases in the amount of care that offspring receive. When
females are widely separated, for example, it may be difficult for males to locate them.
In these cases, males may profit more from helping their current mates and investing
in their offspring than from searching for additional mates. If females are capable of
rearing their offspring alone and need little help from males, then investing males may
be at a reproductive disadvantage compared with males that abandon females after
mating and devote their efforts to finding eligible females.
So far, there is nothing in our reasoning to say that if only one sex invests, it should
In sea horses, always be the female. Why aren’t there primate species in which males do all the work
FIGURE 6.4
males carry fertil- and females compete with each other for access to males? This is not simply a theo-
ized eggs in a special pouch and provide
retical possibility: female sea horses deposit their eggs in their mate’s brood pouch and
care for their young as the young grow.
then swim away and look for a new mate (Figure 6.4). There are whole families of fish
in which male parental care is more common than female parental care; and in several
species of birds—including rheas, spotted sandpipers, and jacanas—females abandon
their clutches after the eggs are laid, leaving their mates to feed and protect the young.
In primates and other mammals, selection tends to favor low male investment
because females lactate and males do not. Among primates, pregnancy and lactation
commit females to invest in their young and limit the benefits of male investment in
offspring. Because offspring depend on their mothers for nourishment during preg-
nancy and after birth, females cannot abandon their young without greatly reducing
the offspring’s chances of surviving. On the other hand, males are never capable of
rearing their offspring without help from females. Therefore, when only one sex invests
in offspring, it is invariably the female. Sometimes males can help females by defend-
ing territories or by carrying infants so that the mother can feed more efficiently, as
siamangs and owl monkeys do. In most cases, however, these benefits are relatively
insignificant, and selection favors males that allocate more time and energy to mating
than to caring for their offspring.
You may be wondering why selection has not produced males that are able to lac-
tate. As we noted in Chapter 3, most biologists believe the developmental changes that
would enable males to lactate would also make them sterile. This is an example of a
developmental constraint.
Pregnancy and lactation are time-consuming and energetically expensive activities for
female primates. The duration of pregnancy plus lactation ranges from 59 days in the
tiny mouse lemur to 255 days in the hefty gorilla. In primates, as in most other ani-
mals, larger animals tend to have longer pregnancies than do smaller animals, but pri-
mates have considerably longer pregnancies than we would expect on the basis of their
body sizes alone. The extended duration of pregnancy in primates is related to the fact
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 153
that brain tissue develops very slowly. Primates have very large brains in relation to
their body sizes, so extra time is needed for fetal brain growth and development dur-
ing pregnancy. Primates also have an extended period of dependence after birth, fur-
ther increasing the amount of care mothers must provide. Throughout this period,
mothers must meet not only their own nutritional requirements but also those of their
growing infants. In some species, offspring may weigh as much as 30% of their
mother’s body weight at the time of weaning.
The energy costs of pregnancy and lactation impose important constraints on
female reproductive behavior. Because it takes so much time and energy to produce an
infant, each female can rear a relatively small number of surviving infants during her
lifetime (Figure 6.5). For example, a female toque macaque who gives birth for the first
time when she is 5 years old and survives to the age of 20 would produce 15 infants
if she gave birth annually and all of her infants survived. This number undoubtedly
represents an upper limit; in the wild, a substantial fraction of all toque macaque
infants die before they reach reproductive age, intervals between successive live births
often last two years, and some females die before they reach old age. Thus, most toque
macaque females will produce a relatively small number of surviving infants over the
course of a lifetime, and each infant represents a substantial proportion of a female’s
lifetime fitness. Therefore, we would expect mothers to be strongly committed to the This female bon-
welfare of their young. FIGURE 6.5
net macaque pro-
duced twins, but only one survived
A female’s reproductive success depends on her ability to obtain enough beyond infancy. Twins are common
resources to support herself and her offspring. among marmosets and tamarins but are
otherwise uncommon among monkeys
and apes. (Photograph by Kathy West.)
In most species of primates, including humans, females must achieve a minimum
nutritional level in order to ovulate and to conceive. For animals living in the wild,
without takeout pizza or 24-hour grocery stores, getting enough to eat each day is
usually a serious challenge. There is considerable evidence that female reproductive
success is limited by the availability of resources within the local habitat. When
females have better access to high-quality resources, they grow faster, mature earlier,
and give birth at shorter intervals. At a number of sites in
Japan, for example, free-ranging monkeys’ natural diets
have been supplemented with wheat, sweet potatoes, rice,
and other foods by humans for many years (Figure 6.6).
This has led to rapid increases in group size (Figure 6.7).
Comparisons of wild and provisioned primates elsewhere
tell a very similar story.
70
60
50
Number of females
40
30
20
10
The size of the Koshima troop
FIGURE 6.7
of Japanese monkeys grew rap- 0
idly when they were provisioned intensively (䊐) and 1950 1955 1960 1965 1970 1975 1980
then dropped when provisioning was restricted (•). Year
duce an infant. These analyses reveal that young females typically reproduce at lower
rates than middle-aged females do. For example, Figure 6.8 shows that young female
baboons and gorillas have lower birth rates than older females do. First-time (primi-
parous) mountain gorilla mothers have 50% higher rates of infant mortality and 20%
longer interbirth intervals after surviving births than older females do. Even among
Japanese macaques, whose diets are enriched by provisioning, young mothers have
longer interbirth intervals than older females do: 67% of first-time mothers skip a year
before producing a second infant, while only 33% of experienced (multiparous) moth-
ers skip years between births.
The relatively low fertility and high infant mortality of young females reflect the
fact that when female monkeys and apes begin to reproduce, they are not yet fully
grown. As a consequence, energetic investment in infants competes directly with ener-
getic investment in their own skeletal growth and development. Younger females also
may lack experience in handling newborn infants, and may not provide appropriate
care for them.
The fertility of females gradually declines as they reach old age (Figure 6.8). The
reasons for this are not well understood but may be linked to the general decline in
function associated with aging and senescence. As females get close to the maximum
life span for their species, they often stop reproducing altogether. For example, the
maximum life span for female Japanese macaques is about 30 years, although few
females live that long; no females reproduce after the age of 25 in this species. Sylvia
Atsalis and Susan Margulis of Chicago’s Lincoln Park Zoo compiled information
about the hormonal profiles of a number of very old captive gorilla females (more than
35 years of age) living in zoos. These old females generally have longer cycles and more
variable hormonal profiles than younger females; some very old females stop cycling
altogether.
Although there is a clear decline in reproductive function in old age, there are
important differences between reproductive termination in monkeys and apes and
menopause in humans. First, there is a gradual decline in fertility with age in other pri-
mates. In humans, there is an abrupt and complete cessation of reproduction at about
50 years of age. Second, relatively few individuals live to the age of reproductive ter-
mination in nonhuman primate populations, while menopause occurs about midway
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 155
Box 6.1
Dominance Hierarchies
In many animal species, ranging from crickets to chickens to Consider the four hypothetical females in Figure 6.9a, which
chimpanzees, competitive encounters within pairs of indi- we shall call Blue, Turquoise, Green, and Lavender. Blue
viduals are common. The outcome of these contests may be always beats Turquoise, Green, and Lavender. Turquoise
related to the participants’ relative size, strength, experience, never beats Blue but always beats Green and Lavender.
or willingness to fight. In many species, for example, larger Green never beats Blue or Turquoise but always beats
and heavier individuals regularly defeat smaller individuals. Lavender. Poor Lavender never beats anybody. We can sum-
If there are real differences in power (based on size, weight, marize the outcome of these confrontations between pairs of
experience, or aggressiveness) between individuals, then we females in a dominance matrix such as the one in Figure
would expect the outcomes of dominance contests to be 6.9b, and we can use the data to assign numerical ranks to
more or less the same from day to day. This is often the case. the females. In this case, Blue ranks first, Turquoise second,
When dominance interactions between two individuals have Green third, and Lavender fourth. When females can defeat
predictable outcomes, we say that a dominance relationship all the females ranked below them and none of the females
has been established. ranked above them, dominance relationships are said to be
When dominance interactions have predictable out- transitive. When the relationships within all sets of three
comes, we can assign dominance rankings to individuals. individuals (trios) are transitive, the hierarchy is linear.
Loser
Defeats _
10 13 6
Defeats _
22 15
Winner
Defeats _ 11
Defeats Nobody _
(a) (b)
(a) Suppose that four hypothetical females—named Blue, Turquoise, Green, and
FIGURE 6.9
Lavender, respectively—have the following transitive dominance relationships: Blue
defeats the other three in dominance contests. Turquoise cannot defeat Blue but can defeat Green and Laven-
der. Green loses to Blue and Turquoise but is able to defeat Lavender. Lavender can’t defeat anyone. (b) The
results of data like those in part a are often tabulated in a dominance matrix, with the winners listed down the
left side and the losers across the top. The value in each cell of the matrix represents the number of times one
female defeated the other. Here, blue defeated Turquoise 10 times and Green defeated Lavender 11 times.
There are no entries below the diagonal because females were never defeated by lower-ranking females.
156
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 157
9
Small group, high rank
8 Small group, low rank
Large group, high rank
7 Large group, low rank
Cumulative reproductive success
0
0 5 8 11 14 17 20 23 26
Female age (years)
(a) (b)
(a) In free-ranging groups of long-tailed macaques, both group size and domi-
FIGURE 6.10
nance rank influence females’ lifetime reproductive success. In general, females
living in small groups reproduce more successfully than do females living in larger groups. But in both
large and small groups, high-ranking females reproduce more successfully than do low-ranking females.
(b) A long-tailed macaque and her nursing infant.
multifemale groups of Old World monkeys, female dominance rank is correlated with
various aspects of females’ reproductive performance. In Amboseli, Kenya, for exam-
ple, the offspring of high-ranking female baboons grow faster and mature earlier than
do the offspring of low-ranking females. In captive vervet groups, high-ranking
females have shorter interbirth intervals than lower-ranking females have. In some
macaque populations, the offspring of high-ranking females are more likely to survive
to reproductive age than the offspring of lower-ranking females. Associations between
dominance rank and reproductive success may produce variation in lifetime fitness
among females, particularly if females maintain the same rank over
the course of their lives, as female macaques and baboons typically
do. Thus, Maria van Noordwijk and Carel van Schaik have found
substantial differences in the lifetime reproductive success of high-,
middle-, and low-ranking long-tailed macaques (Figure 6.10).
In hanuman langurs, female rank also influences female repro-
ductive performance. In this species, female rank is inversely related
to age, so young females typically outrank older ones (Figure 6.11).
Long-term studies of hanuman langurs near Jodhpur, India, con-
ducted by a group of German primatologists, including Carola Bor-
ries (now at Stony Brook, State University of New York) and Volker
Sommer (now at University College London), have shown that
young, high-ranking females reproduce more successfully than do
older, lower-ranking females (Figure 6.12). Studies of hanuman lan-
gurs at Ramnagar in southern Nepal conducted by another group of A female hanuman langur at Jodh-
German primatologists, including Andreas Koenig, Carola Borries, FIGURE 6.11
pur threatens another group mem-
and Paul Winkler, have found that high-ranking females manage to ber. (Photograph courtesy of Carola Borries.)
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 158
0.5
The quality of social bonds may also influence female reproductive success.
1.0
0.9
0.8
Cumulative survival
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 1 2 3 4 5 6 7
Age of offspring (years)
(a)
14
Age of daughter’s maturation (years)
13
12
Among chimpanzees at
11 FIGURE 6.13
Gombe Stream National
Park, Tanzania, female rank influences reproductive
10
performance. (a) The offspring of high- (red line)
and middle-ranking (blue line) females are more
9
likely to survive to weaning age than are the
offspring of low-ranking females (green line).
8
High Middle Low (b) Daughters of high-ranking females also mature
Mother’s rank at daughter’s birth at earlier ages than do the daughters of lower-
(b) ranking females.
0.2
0.0
The sociality index captures information about how much time –0.1
FIGURE 6.15
females spent in close association with other group members.
High values represent more sociable females. More sociable females have higher sur-
–0.2
vivorship among their offspring, and these effects are not due to differences in female
rank or the size of their families. (From Figure 1, in J. B. Silk, S. C. Alberts, and J.
–0.3
Altmann, 2003, “Social Bonds of Female Baboons Enhance Infant Survival,” Science 1 2 3 4
302:1331–1334.) Composite sociality index
others, such as better protection from predators. It’s also possible that social contact
reduces females’ levels of stress, and this has beneficial effects of their health and the
welfare of their offspring.
Reproductive Trade-offs
Females must make a trade-off between the number of offspring that they
produce and the quality of care that they provide.
Just as both males and females must allocate limited effort to parental investment and
mating, females must apportion resources among their offspring. All other things being
equal, natural selection will favor individuals that are able to convert effort into off-
spring most efficiently. Because mothers have a finite amount of effort to devote to off-
spring, they cannot maximize both the quality and the quantity of the offspring they
produce. If a mother invests great effort in one infant, she must reduce her investment
in others. If a mother produces many offspring, she will be unable to invest very much
in any of them.
In nature, maternal behavior reflects this trade-off when a mother modifies her
investment in relation to an offspring’s needs. Initially, infants spend virtually all of
their time in contact with their mothers. The very young infant is entirely dependent
on its mother for food and transportation and is unable to anticipate or to cope with
environmental hazards. At this stage, mothers actively maintain close contact with
their infants (Figure 6.16): retrieving them when they stray too far and scooping them
up when danger arises.
As infants grow older, however, they become progressively more independent and
more competent. They venture away from their mothers to play with other infants and
to explore their surroundings. They begin to sample food plants, sometimes mooching
scraps of their mother’s food. They become aware of the dangers around them, attend-
ing to alarm calls given by other group members and reacting to disturbances within
the group. Mothers use a variety of tactics to actively encourage their infants to
become more independent. They may subtly resist their infants’ attempts to suckle.
A female chim- They may also encourage their infants to travel independently. Nursing is gradually
FIGURE 6.16 limited to brief and widely spaced bouts that may provide the infant with more psy-
panzee sits
beside her youngest infant in Gombe chological comfort than physical nourishment. At this stage, infants are carried only
Stream National Park in Tanzania. when they are ill, injured, or in great danger.
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 161
So far, we have seen that primate females invest heavily in each of their young and pro-
duce relatively few offspring over the course of their lives. Moreover, most primate A large infant
FIGURE 6.18
females can raise their offspring without help from males. Female reproductive suc- squirrel monkey
is being carried by its mother. By the
cess is limited by access to food, not access to mates (Figure 6.19). Males can poten-
time they are weaned, infants in some
tially produce progeny from many females, and as a result, males compete for access
species may weigh up to a third of their
to females. Characteristics that increase male success in competition for mates will mother’s weight. (Photograph by Sue
spread as a result of what Darwin called sexual selection. Boinski.)
It is important to understand the distinction between natural selection and sex-
ual selection. Most kinds of natural selection favor phenotypes in both males and
females that enhance their ability to survive and reproduce. Many of these traits are
related to resource acquisition, predator avoidance, and offspring care. Sexual selec-
tion is a special category of natural selection that favors traits that increase success in
competition for mates, and it will be expressed most strongly in the sex whose access
to members of the opposite sex is most limited. Sexual selection may favor traits that
increase the animal’s attractiveness to potential mates, such as the peacock’s tail, the
red deer’s antlers, and the hamadryas baboon’s mane, even if those traits reduce the
ability of the animal to survive or acquire resources—outcomes not usually favored
by natural selection (Figure 6.20).
(a) (b)
Sexual selection can favor traits not favored by natural selection. (a) The pea-
FIGURE 6.20
cock’s tail hinders his ability to escape from predators, but it enhances his
attractiveness to females. Female peahens are attracted to males that have the most eyespots in their
trains. (b) Male red deer use their antlers when they fight with other males. Red deer antlers are a
good example of a trait that has been favored by sexual selection.
demonstrate that the lifetime reproductive success of the most successful males is often
much greater than that of even the most successful females (Figure 6.21). The same pat-
tern is likely to hold for nonmonogamous primates. A primate male who succeeds in
competition with other males may sire scores of offspring; a successful female might give
(a) (b)
Females
15
The reproduc- Males
FIGURE 6.21
tive success of
male lions (a) is considerably more vari-
10
Breeders (%)
birth to 5 or 10 offspring. Unsuccessful males and females will fail to reproduce at all.
Because the strength of selection depends on how much variation in fitness there is
among individuals, sexual selection acting on male primates can be much stronger than
selective forces acting on female primates. (Incidentally, in species like sea horses, in
which males invest in offspring and females do not, the entire pattern is reversed: sex-
ual selection acts much more strongly on females than on males.)
There are two types of sexual selection: (1) intrasexual selection results from
competition among males, and (2) intersexual selection results from female
choice.
Many students of animal behavior subdivide sexual selection into two categories:
intrasexual selection and intersexual selection. In species in which females cannot
choose their mates, access to females will be determined by competition among
males. In such species, intrasexual selection favors traits that enhance success in
male–male competition. In species in which females can choose the partner(s) with
which they mate, selection favors traits that make males more attractive to females.
This is called intersexual selection. There is not much evidence that intersexual selec-
tion plays an important role in primates, so we will focus our attention on intrasexual
selection.
Intrasexual Selection
Competition among males for access to females favors large body size, large
canine teeth, and other weapons that enhance male competitive ability.
For primates and most other mammals, intrasexual competition is most intense
among males. In the most basic form of male–male competition, males simply drive
other males away from females. Males who regularly win such fights have higher
reproductive success than those who lose. Thus, intrasexual selection favors features
such as large body size, horns, tusks, antlers, and large canine teeth that enable males
to be effective fighters. For example, male gorillas compete
fiercely over access to groups of females, and males weigh
twice as much as females and have longer canine teeth.
As explained in Chapter 5, when the two sexes consis-
tently differ in size or appearance, they are said to be sexu-
ally dimorphic (Figure 6.22). The body sizes of males and
females represent compromises among many competing
selective pressures. Larger animals are better fighters and are
less vulnerable to predation, but they also need more food
and take longer to mature. Intrasexual competition favors
larger body size, larger teeth, and other traits that enhance
fighting ability. Males compete over females; females com-
pete over resources but generally do not compete over mates.
The effect of intrasexual competition among males, however, Adult male baboons are nearly twice the size
is quantitatively greater than the effect of competition among FIGURE 6.22
of adult females. The degree of sexual dimor-
females because the fitness payoff to a successful male is phism in body size is most pronounced in species with the greatest
greater than it is to a successful female. Therefore, sexual competition among males over access to females.
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 164
selection is much more intense than ordinary natural selection. As a result, intrasex-
ual selection leads to the evolution of sexual dimorphism.
The fact that sexual dimorphism is greater in primate species forming one-male,
multifemale groups than in pair-bonded species indicates that intrasexual
selection is the likely cause of sexual dimorphism in primates.
If sexual dimorphism among primates is the product of intrasexual competition
among males over access to females, then we should expect to see the most pro-
nounced sexual dimorphism in the species in which males compete most actively over
access to females. One indirect way to assess the potential extent of competition
among males is to consider the ratio of males to females in social groups. In general,
male competition is expected to be most intense in social groups where males are most
outnumbered by females. At first, this prediction might seem paradoxical, because we
might expect to have more competition when more males are present. The key to
resolving this paradox is to remember that, in most natural populations, there are
approximately equal numbers of males and females at birth. In species that form one-
male groups, there are many bachelor males (males who don’t belong to social
groups) who exert constant pressure on resident males. In species that form pair bonds,
each male is paired with a single female, reducing the intensity of competition among
males over access to females.
Comparative analyses conducted by Paul Harvey of the University of Oxford and
Tim Clutton-Brock of the University of Cambridge have demonstrated that the extent
of sexual dimorphism in primates corresponds roughly to the form of social group in
which the males live (Figure 6.23). There is little difference in body weight or canine
size between males and females in species that typically form pair bonds, such as gib-
bons, titi monkeys, and marmosets. At the other extreme, the most pronounced dimor-
phism is found in species, such as gorillas and black-and-white colobus monkeys, that
live in one-male, multifemale groups. And in species that form multimale, multifemale
groups, the extent of sexual dimorphism is generally intermediate between these
extremes. Thus, sexual dimorphism is most pronounced in the species in which the
ratio of males to females living in bisexual groups is lowest (that is, the relative num-
ber of females is the highest).
1.8 1.8
Female canine length
1.6 1.6
Female body weight
Male canine length
1.4 1.4
The degree of sexual dimor-
FIGURE 6.23
phism is a function of the
ratio of males to females in social groups. Relative 1.2 1.2
canine size (male canine length divided by female
canine length) and body size dimorphism (male body
1.0 1.0
weight divided by female body weight) are greater in
Pair-bonded
species that form one-male, multifemale groups than
One-male, multifemale
in species that form multimale, multifemale groups or
pair-bonded groups. Multimale, multifemale
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 165
In multimale, multifemale groups, where females mate with several males during
2.0
a given estrous period, sexual selection favors increased sperm production.
males, pair-bonded males may do better by guarding their partners when they are sex- Multimale, multifemale
ually receptive than by producing large quantities of sperm. Similarly, competition in
The average
sperm quantity probably does not play an important role in species that form one- FIGURE 6.24
size of testes in
male, multifemale groups. In these species, competition among males is over access to
species that typically form pair-bonded
groups of females, which favors traits related to fighting ability. If resident males are and one-male, multifemale groups is rel-
able to exclude other males from associating with females in their groups, there may atively smaller than the average size of
be little need to produce additional quantities of sperm. testes in multimale, multifemale groups.
Social organization is associated with testis size, much as we would expect. Males Here, observed testes weight is divided
with larger testes typically produce more sperm than do males with smaller testes, and by the expected testes weight to pro-
males who live in multimale groups have much larger testes in relation to their body duce relative testes size. The expected
size than do males who live in either pair-bonded or one-male, multifemale groups testes weight is derived from analyses
(Figure 6.24). Interestingly, in polyandrous groups in which a single breeding female that correct for the effects of body size.
mates regularly with multiple males, males also have relatively large testes.
Investing Males
In species that form pair bonds, males do not compete directly over access to females.
In these species, males’ reproductive success depends mainly on their ability to estab-
lish territories, find mates, and rear surviving offspring. In such pair-bonded species,
mate guarding and offspring care are important components of males’ reproductive
tactics.
Mate guarding may be an important component of pair-bonded males’ repro-
ductive effort. Numerous genetic studies of pair-bonded birds have demonstrated that
a large fraction of the young are not sired by the female’s mate. (This is why biologists
now avoid using the term monogamy to describe pair-bonded species.) Pair-bonded titi
monkeys and gibbons have also been seen copulating with members of neighboring
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 166
80
Grooming (%)
60
40
20
0
Male Female Male Female
(a) (b)
(a) In two white-handed gibbon groups, males groom their mates far more than
FIGURE 6.25
they are groomed in return. The blue bar is the proportion of grooming from
the male to the female, and the red bar is the proportion of grooming from the female to the male.
Males’ solicitous attention to females may be a form of mate guarding. (b) A pair of white-handed gib-
bons (Hylobates lar) grooms.
a year. Males play an active role in child care, frequently carrying infants, 2.5
0
0 1 2 3 4 5 6
Male– Male Competition in Number of adult male/female residents
versity of Oxford has estimated that half of the males involved in aggressive takeover
attempts are seriously wounded. Among hanuman langurs, males form all-male
bands that collectively attempt to oust resident males from bisexual (coed) groups.
Once they succeed in driving out the resident male, the members of the all-male band
compete among themselves for sole access to the group of females. One consequence
of this competition is that male tenure in one-male groups is often short.
Residence in one-male groups does not always ensure exclusive access to females.
For males in multimale groups, conflict arises over group membership and access
to receptive females.
In multimale groups, there is more competition over gaining access to mating part-
ners than over establishing group membership. Nonetheless, it is not necessarily easy
to join a new group; no one puts out the welcome mat. In some macaque species, males
hover near the periphery of social groups, avoid aggressive challenges by resident males,
and attempt to ingratiate themselves with females. In chacma baboons, immigrant
males sometimes move directly into the body of the group
and engage high-ranking resident males in prolonged vocal
duels and chases. Although there may be conflict when males
attempt to join nonnatal groups, males spend most of their
adult lives in groups that contain both males and females.
In multimale groups, males often compete directly over
access to receptive females. Sometimes males attempt to
drive other males away from females, to interrupt copula-
tions, or to prevent other males from approaching or inter-
acting with females. More often, however, male–male
competition is mediated through dominance relationships
that reflect male competitive abilities. These relationships are
generally established in contests that can involve threats and
stereotyped gestures but that can also lead to escalated con-
Male baboons compete over access to an flicts in which males chase, wrestle, and bite one another (Fig-
FIGURE 6.29
estrous female. ure 6.29). Male fighting ability and dominance rank are
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 169
1.2 1
1.0 3
Number of offspring per male
0.6
7
0.4
9
0.2
11
0
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Male age (years)
In baboons, male rank (black dots) is closely linked to male age and physical
FIGURE 6.30
condition. Males reach their highest ranks, on average, when they are 8 to 10
years old, and then gradually fall in rank as they age. Male reproductive success (tan bars) closely corre-
sponds to male age and male rank. (From Figure 2a, in S. C. Alberts, J. C. Buchan, and J. Altmann,
2006, “Sexual Selection in Wild Baboons: From Mating Opportunities to Paternity Success,” Animal
Behaviour 72:1177–1196.)
generally closely linked to physical condition: prime-age males in good physical con-
dition are usually able to dominate others (Figure 6.30).
It seems logical that male dominance rank would correlate with male reproduc-
tive success in multimale groups, but this conclusion has been energetically debated.
The issue has been difficult to resolve because it is difficult to infer paternity from
behavioral observations of mating. However, new genetic techniques make it possible
to assess paternity with a much higher degree of precision.
As more and more genetic information about paternity has become available, the
links between male dominance rank and reproductive success have become stronger.
For example, Susan Alberts, from Duke University, and her colleagues have compiled
information about the reproductive performance of more than 100 nonnatal adult
males who lived in seven baboon social groups over a 13-year period. High-ranking
males sired substantially more offspring than other males (Figure 6.31). Morever, the
highest-ranking males were more likely to mate guard females during cycles in which
they actually conceived than during cycles in which they did not conceive. Among
chimpanzees in the Taï Forest, Christophe Boesch and his colleagues at the Max Planck
Institute for Evolutionary Anthropology have found that the highest-ranking male sires
nearly half of all the infants born in the group (Figure 6.32). Another team from the
same institution, led by Susan Perry, showed that the top-ranking male sired 38% to
70% of all infants born in three multimale groups of white-faced capuchins. Genetic
analyses of paternity in multimale groups of hanuman langurs, long-tailed macaques,
howler monkeys, patas monkeys, and chimpanzees also show that high-ranking
males reproduce more successfully than other males.
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 170
0.35
025
0.20
0.15
0.10
0.05
0
1 2 3 4 5 6 7 8 9 10 11 12 13 Subadult
Male dominance rank
0.6
Observed
0.5 Expected
Proportion of offspring sired
04
0.3
0.2
0.1
0
1 2 3 4 5 6 7 >7
Male social rank
These genetic analyses also reveal that dominance rank is not the only factor that
influences males’ reproductive performance. Among both baboons and chimpanzees,
the highest-ranking male is able to monopolize access to receptive females most effec-
tively when there are relatively few other males present in the group and when there
are relatively few estrus females present at the same time. In most species, males dis-
perse from groups before their own daughters become sexually mature, thus reducing
the likelihood of father-daughter matings. However, in white-faced capuchins, Perry
and her colleagues found that some high-ranking males remain in groups long enough
to potentially mate with their own daughters. Nonetheless, father-daughter matings
are largely avoided. The alpha males sired 79% of the offspring produced by unrelated
females, but only 6% of their own daughters’ offspring.
Infanticide
We have seen that high-ranking males are able to monopolize access to receptive
females, and this generates fierce competition over residence in one-male groups and
competition for high-ranking positions within multimale groups. Sarah Blaffer Hrdy
of the University of California, Davis, was the first to see that these circumstances
might favor the evolution of infanticide as a male reproductive tactic. Her reasoning
was based on the following logic: When a female monkey gives birth to an infant, she
nurses it for a number of months and does not become pregnant again for a consid-
erable period of time. After the death of an infant, however, lactation ends abruptly
and females resume cycling. Thus, the death of nursing infants hastens the resumption
of maternal receptivity. A male who takes over a group or rises to the top-ranking posi-
tion may benefit from killing nursing infants because their deaths cause their mothers
to become sexually receptive much sooner than they would otherwise.
This hypothesis, which has become known as the sexual selection infanticide
hypothesis, was initially controversial. There were no direct observations of males
killng infants, and some researchers found it hard to believe that this form of violence
was an evolved strategy. However, infanticide by males has now been documented in
approximately 40 primate species (and many nonprimate species such as lions).
Researchers have witnessed at least 60 infanticidal attacks in the wild and have
recorded many nonlethal attacks on infants by adult males. There are many more
instances in which healthy infants have disappeared after takeovers or changes in male
rank. Infanticide occurs in species that typically form one-male groups, and in multi-
male groups of savanna baboons, langurs, capuchins, and Japanese macaques.
This body of data enables researchers to test a number of predictions derived from
Hrdy’s hypothesis. If infanticide is a male reproductive strategy, then we would
expect that (1) infanticide would be associated with changes in male residence or sta-
tus; (2) males should kill infants whose deaths hasten their mother’s resumption of
cycling; (3) males should kill other males’ infants, not their own; and (4) infanticidal
males should achieve reproductive benefits.
All of these predictions have been supported. Carel van Schaik recently compiled
information about 55 infanticides in free-ranging groups that were actually seen by
observers. He found that nearly all infanticides (85%) followed changes in male res-
idence or dominance rank. He also found that most infanticides involve unweaned
06_Primate Mating [3p]:How Humans Evolved [5e] 9/26/08 8:02 PM Page 172
Although infanticide may enhance male reproductive success, it can have a dis-
astrous effect on females who lose their infants. Thus, we should expect females
to evolve counterstrategies to infanticidal threats. The most obvious counterstrat-
egy would be for females to try to prevent males from harming their infants. How-
ever, females’ efforts to defend their infants are unlikely to be effective. Remember
that males are generally larger than females in species without pair bonds, and the
extent of sexual dimorphism is most pronounced in species that form one-male
groups.
Females may try to confuse males about paternity. As we
have seen already, males seem to kill infants when there is no
ambiguity about their paternity; if females can increase
uncertainty about paternity, they may reduce the risk of
infanticide. In some primate species, females mate with a
number of different males while they are receptive; and in
some species, females continue to mate during pregnancy.
Both behaviors may reduce certainty about paternity, thereby
reducing the risk of infanticide.
FURTHER READING
Altmann, J. 1980, 2001. Baboon Mothers and Infants. Chicago: University of Chicago
Press.
Kappeler, P. M., C. P. van Schaik, eds. 2004. Sexual Selection in Primates: New and
Comparative Perspectives. New York: Cambridge University Press.
Krebs, J. R., and N. B. Davies. 1993. An Introduction to Behavioural Ecology, 3rd ed.
Boston: Blackwell Scientific. chs. 8 and 9.
Maestripieri, D., ed. 2003. Primate Psychology. Cambridge, Mass.: Harvard Univer-
sity Press.
Swedell, L. 2006. Strategies of Sex and Survival in Hamadryas Baboons: Through a
Female Lens. Upper Saddle River, N.J.: Pearson Prentice Hall.
van Schaik, C. P., and C. H. Janson, eds. 2000. Infanticide by Males and Its Implica-
tions. New York: Cambridge University Press.
KEY TERMS
Chapter 7
Altruism: A Puzzle
Mutualism
The Problem with Group-Level
Explanations
Kin Selection
Hamilton’s Rule
Kin Recognition
Kin Biases in Behavior
Parent–Offspring Conflict
Reciprocal Altruism
ALTRUISM: A PUZZLE
S
o far, we have explained the evolution of morphology and behavior in terms of
individual reproductive success. Natural selection favored deeper beaks in Dar-
win’s finches during the drought because deeper beaks allowed individuals to
crack tougher seeds. It favors infanticide by males because it allows them to sire off-
spring with the dead infants’ mothers. However, primates (and many other creatures)
perform altruistic behaviors that benefit others but at a personal cost. For example,
virtually all social primates groom other group members, removing parasites, clean-
ing scabs, and picking bits of debris from their hair (Figure 7.1). Grooming other
176
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 177
Mutualism 177
MUTUALISM
Sometimes helping others benefits the actor as well as the
recipient. Such behaviors are mutualistic.
Two male
FIGURE 7.3
baboons (on the
right) jointly challenge a third male over
access to a receptive female.
females are most likely to conceive, the highest-ranking males usually attempt to
monopolize access to females. Two lower-ranking males may jointly challenge a
higher-ranking male and try to gain control of the female he is guarding (Figure 7.3).
These interactions can escalate to energetically costly chases and physical confronta-
tions. The challengers often succeed in driving the higher-ranking male away, and one
of them begins to mate-guard the female. Males who hold middle-ranking positions
are most likely to form coalitions. These males have very little chance of gaining access
to receptive females on their own, but two middle-ranking males are a formidable
force when they work together. As long as each male has some probability of ending
up with the female, it may be profitable for both to participate in the coalition. There
is no incentive to slack off, because doing so guarantees failure.
Like male baboons, chimpanzees also mate-guard receptive females. In the Kibale
Forest, some pairs of high-ranking males jointly defend access to females. Coopera-
tion allows males to fend off other males’ approaches and keep close tabs on females
at the same time. Coalition partners share matings with the female that they are guard-
ing. Males switch from mate guarding alone to joint mate guarding when they are in
large parties with many potential competitors. Again, this strategy may pay off for
males if they are able to obtain more matings when they mate-guard with other males
than when they attempt to monopolize females on their own.
You might think that if the average effect of an act on all members of the group is pos-
itive, then it would be beneficial for all individuals to perform it. For example, suppose
that when one monkey gives an alarm call, the other members of the group benefit, and
the total benefits to all group members exceed the cost of giving the call. Then, if every
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 179
individual gave the call when a predator was sighted, all members of the group would
be better off than if no warning calls were ever given. Since every individual benefits,
many people think that alarm calling will be favored by natural selection.
This inference is wrong because it confuses the effect on the group with the effect
on the actor. In most circumstances, the fact that alarm calls are beneficial to those
hearing them doesn’t affect whether the trait of alarm calling evolves; all that matters
is the effect that giving the alarm call has on the caller. To see why this is true, imag-
ine a hypothetical monkey species in which some individuals give alarm calls when
they are the first to spot a predator. Monkeys who are warned by the call have a chance
to flee. Suppose that one-fourth of the population (“callers”) give the call when they
spot predators, and three-fourths of the individuals (“noncallers”) do not give an
alarm call in the same circumstance. (These proportions are arbitrary; we chose them
because it’s easier to follow the reasoning in examples with concrete numbers.) Let’s
suppose that in this species the tendency to give alarm calls is genetically inherited.
Now we compare the fitness of callers and noncallers. Because everyone in the
group can hear the alarm calls and take appropriate action, alarm calls benefit every-
one in the group to the same extent (Figure 7.4). Calling has no effect on the relative
fitness of callers and noncallers because, on average, one-fourth of the beneficiaries will
be callers and three-fourths of the beneficiaries will be noncallers—the same propor-
tions we find in the population as whole. Calling reduces the risk of mortality for
everyone who hears the call, but it does not change the frequency of callers and non-
callers in the population, because everyone gains the same benefits. However, callers
are conspicuous when they call, so they are more vulnerable to predators. Although
all individuals benefit from hearing alarm calls, callers are the only ones who suffer
the costs from calling. This means that, on average, noncallers will have a higher fit-
ness than callers. Thus, genes that cause alarm calling will not be favored by selection,
+ –
–
+ –
+
–
– + –
+ –
+
+ – –
+
(a) Altruist gives alarm call to group. (b) Nonaltruist doesn't give alarm call to group.
Two groups of monkeys are approached by a predator. (a) In one group, one indi-
FIGURE 7.4
vidual (pink) has a gene that makes her call in this context. Giving the call lowers
the caller’s fitness but increases the fitness of every other individual in the group. Like the rest of the
population, one out of four of these beneficiaries also carries the genes for calling. (b) In the second
group, the female who detects the predator does not carry the gene for calling and remains silent. This
lowers the fitness of all members of the group a certain amount because they are more likely to be
caught unaware by the predator. Once again, one out of four is a caller. Although members of the
caller’s group are better off on average than members of the noncaller’s group, the gene for calling is
not favored, because callers and noncallers in the caller’s group both benefit from the caller’s behavior
but callers incur some costs. Callers are at a disadvantage relative to noncallers. Thus, calling is not
favored, even though the group as a whole benefits.
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 180
Box 7.1
Group Selection
Group selection was once thought to be the mechanism for can occur if groups vary in their ability to survive and to
the evolution of altruistic interactions. In the early 1960s, reproduce and if that variation is heritable. Then group
British ornithologist V. C. Wynne-Edwards contended that selection may increase the frequency of genes that increase
altruistic behaviors like those we have been considering here group survival and reproductive success. The strength of
evolved because they enhanced the survival of whole groups selection among groups depends on the amount of genetic
of organisms. Thus, individuals gave alarm calls, despite the variation among groups, just as the strength of selection
costs of becoming more conspicuous to predators, because among individuals depends on the amount of genetic vari-
calling protected the group as a whole from attacks. Wynne- ation among individuals. However, when individual selec-
Edwards reasoned that groups that contained a higher num- tion and group selection are opposed and group selection
ber of altruistic individuals would be more likely to survive favors altruistic behavior while individual selection favors
and prosper than groups that contained fewer altruists, selfish alternatives, individual selection has a tremendous
and the frequency of the genes leading to altruism would advantage. This is because the amount of variation among
increase. groups is much smaller than the amount of variation among
Wynne-Edwards’s argument is logical because Darwin’s individuals, unless groups are very small and there is very
postulates logically apply to groups as well as individuals. little migration among them. Thus, individual selection
However, group selection is not an important force in favoring selfish behavior will generally prevail over group
nature, because there is generally not enough genetic varia- selection, making group selection an unlikely source of
tion among groups for selection to act on. Group selection altruism in nature.
even if the cost of giving alarm calls is small and the benefit to the rest of the group is
large. Instead, selection will favor genes that suppress alarm calling, because noncallers
have higher fitness than callers. (See Box 7.1 for more about the role of group selec-
tion in nature.)
KIN SELECTION
Natural selection can favor altruistic behavior if altruistic individuals are more
likely to interact with each other than chance alone would dictate.
therefore, callers and noncallers are equally likely to benefit from hearing an alarm call.
Hamilton’s insight was to see that any process that causes altruists to be more likely
to interact with other altruists than they would by chance could facilitate the evolu-
tion of altruism.
To see why this is such an important insight, let’s modify the previous example by
assuming that our hypothetical species lives in groups composed of full siblings, off-
spring of the same mother and father. The frequencies of the calling and noncalling
genes don’t change, but their distribution will be affected by the fact that siblings live
together (Figure 7.5). If an individual is a caller, then, by the rules of Mendelian genet-
ics, there is a 50% chance that the individual’s siblings will share the genes that cause
calling behavior. This means that the frequency of the genes for calling will be higher
in groups that contain callers than in the population as a whole and, therefore, more
than one-fourth of the beneficiaries will be callers themselves. When a caller gives an
alarm call, the audience will contain a higher fraction of callers than the population
at large does. Thus, the caller raises the average fitness of callers relative to noncallers.
Similarly, because the siblings of noncallers are more likely to be noncallers than
chance alone would dictate, callers are less likely to be present in such groups than in
the population at large. Therefore, the absence of a warning call lowers the relative
fitness of noncallers relative to callers.
When individuals interact selectively with relatives, callers are more likely to ben-
efit than noncallers and, all other things being equal, the benefits of calling will favor
the genes for calling. However, we must remember that calling is costly, and this will
+ –
–
+ –
+
–
– + –
+ –
+
+ – –
+
(a) Altruist gives alarm call to siblings. (b) Nonaltruist doesn't give alarm call to siblings.
tend to reduce the fitness of callers. Calling will be favored by natural selection only
if its benefits are sufficiently greater than its costs. The exact nature of this trade-off
is specified by what we call “Hamilton’s rule.”
Hamilton’s Rule
Hamilton’s theory of kin selection is based on the idea that selection could favor altru-
istic alleles if animals interacted selectively with their genetic relatives. Hamilton’s the-
ory also specifies the quantity and distribution of help among individuals. According
to Hamilton’s rule, an act will be favored by selection if
rb ⬎ c
where
r = the average coefficient of relatedness between the actor and the recipients
b = the sum of the fitness benefits to all individuals affected by the behavior
c = the fitness cost to the individual performing the behavior
TABLE 7.1 The value of r for selected categories of relatives. (When cousins are off-
spring of full siblings, they are related by 0.125, but when they are the off-
spring of half-siblings they are related by 0.0625.)
Relationship r
bility that both siblings will acquire the same allele from their mother is still 0.25, but
female B and male C might also share an allele from their father. The probability of
this event is also 0.25. Thus the probability that female B and male C share an allele
is equal to the sum of 0.25 and 0.25, or 0.5. This basic reasoning can be extended to
calculate the degrees of relatedness among various categories of kin (Table 7.1).
Hamilton’s rule leads to two important insights: (1) altruism is limited to kin,
and (2) closer kinship facilitates more costly altruism.
If you reflect on Hamilton’s rule for a while, you will see that it produces two fun-
damental predictions about the conditions that favor the evolution of altruistic behav-
iors. First, altruism is not expected to be directed toward nonkin, because the
coefficient of relatedness, r, between nonkin is 0. The condition for the evolution of
altruistic traits will be satisfied only for interactions between kin, when r ⬎ 0. Thus,
altruists are expected to be nepotistic, showing favoritism toward kin.
Second, close kinship is expected to facilitate altruism. If an act is particularly
costly, it is most likely to be restricted to close kin. Figure 7.7 shows how the benefit/
cost ratio scales with the degree of relatedness among individuals. Compare what hap-
140
120
Benefit/cost ratio, b/c
100
80
60
As the
FIGURE 7.7
degree of
40 relatedness (r) between two indi-
viduals declines, the value of the
20 ratio of benefits to costs (b/c) that
0 is required to satisfy Hamilton’s rule
0.5 0.25 0.125 0.0625 0.03125 0.0156 0.0078 for the evolution of altruism rises
Coefficient of relatedness, r rapidly.
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 184
pens when r = 1⁄16 (or 0.0625) and when r = 1⁄2 (or 0.5). When r = 1⁄16, the benefits must
be more than 16 times as great as the costs for Hamilton’s inequality rb ⬎ c to be sat-
isfied. When r = 1⁄2, the benefit needs to be just over twice as large as the costs. All other
things being equal, altruism will be more common among close relatives than among
distant ones.
Kin Recognition
In order for kin selection to provide an effective mechanism for the evolution of coop-
erative behavior, animals must be able to distinguish relatives from nonrelatives and
close relatives from distant ones. Some organisms are able to recognize their kin by
their smell or likeness to themselves. This is called phenotypic matching. Others learn
to recognize relatives using contextual cues—cues such as familiarity and proximity—
that predict kinship (Figure 7.8). We have generally assumed that primates rely on con-
textual cues to identify their relatives, but new data suggest that phenotypic matching
may also play a role in primate kin recognition.
Mothers seem to make use of contextual cues to recognize their own infants. After
they give birth, females repeatedly sniff and inspect their newborns. By the time their
infants are a few weeks old, mothers are clearly able to recognize them. After this,
females of most species nurse only their own infants and respond selectively to their
own infants’ distress calls. Primate mothers don’t really need innate means of recog-
nizing their young, because young infants spend virtually all of their time in physical
contact with their mothers. Thus, mothers are unlikely to confuse their own newborn
with another.
Monkeys and apes may learn to recognize other maternal kin through contact
with their mothers. Offspring continue to spend considerable amounts of time with
their mothers even after their younger siblings are born. Thus, they have many oppor-
tunities to watch their mothers interact with their new brothers and sisters. Similarly,
the newborn infant’s most common companions are its mother and siblings (Fig-
ure 7.9). Because adult females continue to associate with their mothers, infants also
become familiar with their grandmothers, aunts, and cousins.
Contextual cues may play some role in paternal kin recognition as well.
Primatologists were once quite confident that primates were unable to recognize
their paternal kin. This conclusion was based on the following reasoning: First, pair
bonds are uncommon in most primate species, so patterns of association between
males and females would not provide accurate cues of paternal kinship. Second,
females may mate with several different males near the time of conception, creating (a)
confusion about paternity. Even in pair-bonded species, like gibbons and Callicebus
monkeys, females sometimes mate with males from outside their groups. However, as
we explained in Chapter 6, male baboons are somehow able to identify their own off-
spring. This tells us that conventional wisdom is sometimes wrong.
New evidence from field studies on macaques and baboons suggests that monkeys
use contextual cues to assess paternal kinship. Jeanne Altmann of Princeton Univer-
sity pointed out that age may provide a good proxy measure of paternal kinship in
species in which a single male typically dominates mating activity within the group.
When this happens, all infants born at about the same time are likely to have the same
father. Recent studies suggest that Altmann’s logic is correct—that is, that monkeys
do use age to identify paternal kin. Anja Widdig of the Max Planck Institute for Evo- (b)
lutionary Anthropology in Leipzig, Germany, and her colleagues studied kin recogni-
Monkey and
tion among rhesus macaques on Cayo Santiago, Puerto Rico. In this population, a FIGURE 7.9
ape infants grow
single male dominated matings during each mating season. Females showed strong up surrounded by various relatives.
affinities for maternal half sisters, as expected. However, females also showed strong (a) These adult baboon females are
affinities for their paternal half sisters, spending more time grooming them and more mother and daughter. Both have young
time in close proximity to them than to unrelated females (Figure 7.10). Their affini- infants. (b) An adolescent female bon-
ties for paternal kin seem to be based partly on strong preferences for interacting with net macaque carries her younger brother
age-mates. However, females also distinguished among their age-mates, preferring while her mother recovers from a
paternal half sisters over unrelated females of the same age. Female baboons in serious illness.
0.3
0.2
Female macaques are able to identify
FIGURE 7.10
paternal siblings. Females interact far more
often with maternal half siblings (blue bar) than with paternal sib-
0.1 lings, but they interact more often with paternal siblings than with
nonkin. Age similarity seems to provide a cue for paternal kinship:
females interact more often with paternal half-sibling peers (red
hatched bar) than with paternal half siblings that are not close in
0 age (solid red bar). But note that females also distinguish among
Maternal Peer Nonpeer Peer Nonpeer peers, preferring half-sibling peers over unrelated peers (purple
siblings
Paternal siblings Nonkin hatched bar).
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 186
Amboseli, Kenya, also distinguish between paternal half sisters and unrelated females,
but it is not clear whether they can distinguish paternal half sisters from unrelated age-
mates.
Social grooming plays an important role in the lives of most gregarious primates
(Figure 7.11). Grooming is likely to be beneficial to the participants in at least two
ways: First, grooming serves hygienic functions because bits of dead skin, debris, and
parasites are removed and wounds are kept clean and open. Second, grooming may
provide a means for individuals to establish relaxed, affiliative (friendly) contact and
to reinforce social relationships with other group members (Box 7.2). Grooming is
costly because the actor expends both time and energy in performing these services.
Moreover, Marina Cords of Columbia University has shown that blue monkeys are
less vigilant when they are grooming, perhaps exposing themselves to some risk of
being captured by predators.
Grooming is more common among kin, particularly mothers and their offspring,
than among nonkin. For example, Ellen Kapsalis and Carol Berman of the University
(a) (b)
Some of the
FIGURE 7.11
many species
of primates that groom are (a) capuchin
monkeys, (b) blue monkeys, (c)
baboons, and (d) gorillas. (Photographs
courtesy of Susan Perry (a), Marina
Cords (b), and John Mitani (d).)
(c) (d)
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 187
Box 7.2
How Relationships
Are Maintained
Conflict and competition are fundamental features of social sometimes exchanged for support in coalitions, and groom-
life for many primates: females launch unprovoked attacks ing partners may be allowed to share access to scarce
on unsuspecting victims, males battle over access to recep- resources.
tive females, subordinates are supplanted from choice feed- When tensions do erupt into violence, certain behav-
ing sites, and dominance relationships are clearly defined ioral mechanisms may reduce the disruptive effects of con-
and frequently reinforced. Although violence and aggression flict on social relationships. After conflicts end, victims often
are not prevalent in all primates (muriquis, for example, are flee from their attackers—an understandable response. In
so peaceful that dominance hierarchies cannot be detected), some cases, however, former opponents make peaceful con-
some primates can be charitably characterized as con- tact in the minutes that follow conflicts. For example, chim-
tentious. This raises an intriguing question: how is social life panzees sometimes kiss their former opponents, female
sustained in the face of such relentless conflict? After all, it baboons grunt quietly to their former victims, and golden
seems inevitable that aggression and conflict will drive ani- monkeys may embrace or groom their former adversaries.
mals apart, disrupt social bonds, and reduce the cohesive- The swift transformation from aggression to affiliation
ness of social groups. Recently, primatologists have begun to prompted Frans de Waal of Emory University to suggest that
consider this issue. these peaceful postconflict interactions are a form of recon-
Social relationships matter to primates. They spend a ciliation, a way to mend relationships that were damaged by
considerable portion of every day grooming other group conflict. Inspired by de Waal’s work, a number of research-
members. Grooming is typically focused on a relatively ers have documented reconciliatory behavior in a wide
small number of partners and is often reciprocated. Robin range of primate species.
Dunbar of the University of Oxford contends that in Old Peaceful postconflict interactions seem to have a calm-
World monkeys, grooming has transcended its original ing effect on former opponents. When monkeys are nervous
hygienic function and now serves as a means to cultivate and anxious, rates of certain self-directed behaviors, such as
and maintain social bonds. Social bonds may have real scratching, increase. Thus, self-directed behaviors are a good
adaptive value to individuals. For example, grooming is behavioral index of stress. Filippo Aureli of Liverpool John
0.60
187
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 188
Moores University and his colleagues at Utrecht University some researchers have argued that kin have little need to rec-
and Emory University have studied the effects of fighting and oncile, because their relationships are unlikely to be frayed
reconciliation on the rate of self-directed behaviors. They by conflict.
have found that levels of self-directed behavior, and pre- Reconciliation may also play a role in resolving con-
sumably stress, rise above baseline levels after conflicts. flicts among individuals who do not have strong social
Both victims and aggressors seem to feel the stressful effects bonds. Like many other primates, female baboons are
of conflicts. If former opponents interact peacefully in the strongly attracted to newborn infants and make persistent
minutes that follow conflicts, rates of self-directed behavior efforts to handle them. Mothers reluctantly tolerate infant
fall rapidly to baseline levels (Figure 7.12). If they do not rec- handling, but they do not welcome the attention. Female
oncile, rates of self-directed behavior remain elevated above baboons reconcile at particularly high rates with the moth-
baseline levels for several minutes longer. If reconciliation ers of young infants, even when they do not have close rela-
provides a means to preserve social bonds, then we would tionships with them. Reconciliation greatly enhances the
expect primates to reconcile selectively with their closest likelihood that aggressors will be able to handle their former
associates. In a number of groups, former opponents who victims’ infants in the minutes that follow conflicts. Thus, in
have strong social bonds are most likely to reconcile. Kin this case, reconciliation seems to be a means to an immedi-
also reconcile at high rates in some groups, even though ate end but not a means to preserve long-term relationships.
Most disputes in primate groups involve two individuals. Sometimes, however, sev-
eral individuals jointly attack another individual, or one individual comes to the sup-
port of another individual involved in an ongoing dispute (Figure 7.14). We call these
kinds of interactions coalitions or alliances. Support is likely to be beneficial to the indi-
vidual who receives aid because support alters the balance of power among the origi-
25
Time spent grooming (%)
20
15
10
3.0
Close kin
2.5
Distant kin
Acts per partner
2.0
Nonkin In five groups of wild baboons, rates
FIGURE 7.15
1.5 of coalitionary support provided by
close kin (mothers, daughters, and sisters), more distant
1.0 maternal kin, and all others are shown. In all five groups, sup-
port is biased toward close kin. (From Figure 1, in J. B. Silk,
0.5 S. C. Alberts, and J. Altmann, “Patterns of Coalition Forma-
tion by Adult Female Baboons in Amboseli, Kenya,” Animal
0
Linda’s Nyano’s Omo’s Viola’s Weaver’s Behaviour 67:573–582.)
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 190
Low 30
25
20
Juvenile rank
15
10 (b)
influences the duration and stability of male coalitions. Coalitions that are made up
of related males last nearly four times as long as coalitions composed of unrelated
males. Coalitions composed of related males are also less likely to experience rank
reversals. In this case, the costs of cooperation may be balanced by gains in inclusive
fitness.
Parent–Offspring Conflict
Kin selection helps to explain why there is conflict between parents and
offspring and among siblings.
As we explained in the last chapter, mothers must wean their infants so that they can
conserve energy for subsequent infants. As mothers begin to curtail investment, their
infants often resist, sometimes vigorously. Chimpanzee infants throw full-fledged
tantrums when their mothers rebuff their efforts to nurse, and baboons whimper
piteously when their mothers refuse to carry them. These weaning conflicts arise from
a fundamental asymmetry in the genetic interests of mothers and their offspring.
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 192
Mothers are equally related to all of their offspring (r = 0.5), but offspring are more
closely related to themselves (r = 1.0) than to their siblings (r = 0.5 or 0.25). This phe-
nomenon was labeled parent–offspring conflict by Rutgers biologist Robert Trivers,
who was the first to recognize the evolutionary rationale underlying the conflict
between parents and their offspring.
To understand why there is parent–offspring conflict, imagine a mutation that
increases the amount of maternal investment in the current infant a small amount,
thereby reducing investment in future infants by the same amount. According to
Hamilton’s rule, selection will favor the expression of this gene in mothers if:
Because the mother shares half of her genes with each of her offspring, 0.5 appears
on both sides of the inequality. The inequality tells us that selection will increase invest-
ment in the current offspring until the benefits to the current offspring are equal to the
costs to future offspring. The result is quite different if the genes expressed in the cur-
rent infant control the amount of maternal investment. This time, consider a gene that
is expressed in the current infant that increases the investment the infant receives by
a small amount. Once again, we use Hamilton’s rule, this time from the perspective
of the current infant:
In this case, the infant is related to itself by 1.0 and to its full sibling by 0.5. Now
selection will increase the amount of maternal investment until the incremental ben-
efit of another unit of investment in the current infant is twice the cost to future broth-
ers and sisters of the fetus (and four times for half siblings). Thus, genetic asymmetries
lead to a conflict of interest between mothers and their children. Selection will favor
mothers that provide less investment than their infants desire, and selection will favor
children that demand more investment than their mothers are willing to give. This con-
flict of interest plays out in weaning tantrums and sibling rivalries.
RECIPROCAL ALTRUISM
Altruism can also evolve if altruistic acts are reciprocated.
The theory of reciprocal altruism relies on the basic idea that altruism among indi-
viduals can evolve if altruistic behavior is balanced between partners (pairs of inter-
acting individuals) over time. In reciprocal relationships, individuals take turns being
actor and recipient—giving and receiving the benefits of altruism (Figure 7.18).
Reciprocal altruism is favored because over time the participants in reciprocal acts
obtain benefits that outweigh the costs of their actions. This theory was first formu-
lated by Robert Trivers of Rutgers, and later amplified and formalized by others.
Three conditions occurring together favor the development of reciprocal altruism:
individuals must (1) have an opportunity to interact often, (2) have the ability to keep
track of support given and received, and (3) provide support only to those who help
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 193
them. The first condition is necessary so that individuals will have the opportunity
for their own altruism to be reciprocated. The second condition is important so that
individuals can monitor and balance altruism given to and received from particular
partners. The third condition produces the nonrandom interaction necessary for the
evolution of altruism. If individuals are unrelated, initial interactions will be randomly
distributed to altruists and nonaltruists. However, reciprocators will quickly stop help-
ing those who do not help in return, while continuing to help those who do. Thus, as
in the case of kin selection, reciprocal altruism can be favored by natural selection
because altruists receive a disproportionate share of the benefits of altruistic acts. Note
that altruistic acts need not be exchanged in kind; it is possible for one form of altru-
ism (such as grooming) to be exchanged for another form of altruism (such as coali-
tionary support).
In primates, the conditions for the evolution of reciprocal altruism probably are
satisfied often, and there is some evidence that it occurs.
Most primates live in social groups that are fairly stable, and they can recognize
all of the members of their groups. We do not know whether primates have the cog-
nitive capacity to keep track of support given and received from various partners, but
we do know that they are very intelligent and can solve complex problems. Thus, pri-
mates provide a good place to look for examples of contingent forms of reciprocity.
Reciprocal grooming and coalitionary support have been observed in several
species of macaques, baboons, vervet monkeys, and chimpanzees. In some cases,
grooming and support are exchanged in kind; in other cases, monkeys seem to
exchange grooming for support. Some monkeys switch roles during grooming bouts
so that the amount of grooming given and received during each grooming bout is bal-
anced; others balance grooming across bouts.
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 194
After grooming Among male chimpanzees, social bonds seem to be based on reciprocal exchanges
10
No prior grooming in many different currencies (Figure 7.18). For example, John Mitani and David Watts
have found that male chimpanzees at Ngogo, a site in the Kibale Forest of Uganda,
Duration of response (sec)
share meat selectively with males who share meat with them and with males who reg-
ularly support them in agonistic interactions. Males who hunt together also tend to
groom one another selectively, support one another, and participate in border patrols
5 together. Interestingly, close associates are often not maternal or paternal kin, sug-
gesting that males’ relationships are based on reciprocity, not kinship.
These correlational findings are consistent with predictions derived from the the-
ory of reciprocal altruism, but they do not demonstrate that altruism is contingent on
reciprocation.
0 Kin Nonkin
A few studies suggest that primates keep track of these contingencies, at least
Vervet mon-
FIGURE 7.19
keys responded
over short periods.
more strongly to recruitment calls
played from a hidden speaker if the Robert Seyfarth and Dorothy Cheney conducted the first study to examine the con-
caller had previously groomed them than tingent nature of altruistic exchanges. Like most other monkeys, vervets spend much
if the caller had not. of their free time grooming. Vervets also form coalitions and use specific vocalizations
to recruit support. In this experiment, Seyfarth and Cheney played tape-recorded
recruitment calls to individuals in two different situations. Vervet A’s recruitment call
was played to vervet B from a hidden speaker (1) after A had groomed B and (2) after
a fixed period of time in which A and B had not groomed. It was hypothesized that if
grooming were associated with support in the future, then B should respond most
strongly to A’s recruitment call after being groomed. And that’s just what the vervets
did (Figure 7.19).
Frans de Waal of Emory University designed another study, which focused on the
relationship between grooming and food sharing in a well-established group of cap-
tive chimpanzees. At various times over a three-year period, a group of chimpanzees
was provided with compact bundles of leaves. Although these bundles could be
monopolized by a single individual, possessors of these bundles often allowed other
individuals to share some of their booty. Before each provisioning session, de Waal
monitored grooming interactions among group members. He found that the posses-
sors were more generous to individuals who had recently groomed them than they
were toward other group members (Figure 7.20). Moreover, possessors were less likely
90%
Mean food-getting success
80%
1.0
0.8
Frequency of pulls
0.6
0.4
0.2
0
1 2 3 4
Session
(a) (b)
(a) The actor (on the right) pulls the bar and delivers a food reward to the
FIGURE 7.21
recipient (on the left). The actor and recipient are separated by a transparent
partition. (b) Monkeys were more likely to pull and deliver food rewards to partners who always pulled
for them (red bars) than to partners who never pulled for them (blue bars). (From Figure 2, in M. D.
Hauser, M. K. Chen, F. Chen, and E. Chuang, 2003, “Give unto Others: Genetically Unrelated Cot-
ton-Tamarin Monkeys Preferentially Give Food to Those Who Altruistically Give Food Back,” Pro-
ceedings of the Royal Society of London, Series B 270:2363–2370.)
to actively resist efforts to obtain food if their solicitors had recently groomed them
than if they had not groomed them.
Marc Hauser and his colleagues at Harvard University conducted an experiment
to examine tamarins’ capacity for contingent reciprocity. The tamarins pulled a bar
that delivered a food reward to another tamarin in an adjacent cage but not to the
actor. He trained two monkeys as confederates; one always pulled, and the other never
pulled. When these two confederates were paired with naïve subjects, the trajectory
of behavior differed substantially. Monkeys who were paired with the confederates
who always pulled were more likely to pull in return than monkeys who were paired
with the confederates that never pulled (Figure 7.21). This suggests that the tamarins
were sensitive to the behavior of their partners, as required for contingent reciprocity
to evolve.
The number of well-documented examples of reciprocal altruism in nonhuman
primates is still small. It is possible that reciprocity is actually uncommon in nature.
However, reciprocal altruism may occur more often than is detected by observers.
Because altruism is potentially reciprocated in different currencies (grooming for sup-
port, predator defense for food, and so on), and because the actual costs and benefits
associated with specific behaviors are hard to quantify, it is very difficult to establish
whether altruism is actually reciprocated.
FURTHER READING
Chapais, B., and C. M. Berman, eds. 2004. Kinship and Behavior in Primates. New
York: Oxford University Press.
Dugatkin, L. A. 1997. Cooperation among Animals: An Evolutionary Perspective.
New York: Oxford University Press.
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 196
Harcourt, A. H., and F. B. M. de Waal, eds. 1992. Coalitions and Alliances in Humans
and Other Animals. New York: Oxford University Press.
Kappeler, P. M., and C. P. van Schaik, eds. 2006. Cooperation in Primates and
Humans: Mechanisms and Evolution. New York: Springer.
Krebs, J. R., and N. B. Davies. 1993. An Introduction to Behavioural Ecology. 3rd ed.
Boston: Blackwell Scientific. Ch. 11.
Silk, J. B. 2007. The strategic dynamics of cooperation in primate groups. Advances
in the Study of Behavior 37:1–42.
1. Consider the kinship diagram shown here. What is the kinship relationship (for
example, mother, aunt, or cousin) and degree of relatedness (such as 0.5 or 0.25)
for each pair of individuals?
B C
E D
2. In biological terms, what is the difference between the following situations: (a) A
male monkey sitting high in a tree gives alarm calls when he sees a lion at a dis-
tance. (b) A female monkey abandons a desirable food patch when she is
approached by another female.
3. In documentaries about animal behavior, animals are often said to do things “for
the good of the species.” For example, when low-ranking animals do not repro-
duce, they are said to give up reproducing in order to prevent the population from
becoming too numerous and exhausting its resource base. What is wrong with this
line of reasoning?
4. Why is some sort of nonrandom interaction among altruists necessary for altru-
ism to be maintained?
5. Suppose that primates are not able to recognize paternal kin, as many primatol-
ogists have assumed. What would that tell you about how natural selection pro-
duces adaptations?
6. Data from a number of studies indicate that primates are more likely to behave
altruistically toward kin than toward nonkin. However, many of the same stud-
07_Evolution of Cooperation [3p]:How Humans Evolved [5e] 9/26/08 8:33 PM Page 197
ies show that rates of aggression toward kin and nonkin are basically the same.
How does this fit with what you have learned about kin selection? Why are mon-
keys as likely to fight with kin as with nonkin?
7. There are relatively few good examples of reciprocal altruism in nature. Why is
reciprocal altruism uncommon? Why might we expect reciprocal altruism to be
more common among primates than among other kinds of animals?
8. Seyfarth and Cheney found that vervet monkeys tended to respond more strongly
to the calls of the animals who had groomed them earlier in the day than to the
calls of animals who had not groomed them. However, this effect held only for
unrelated animals, not for kin. The vervets responded as strongly to the calls of
grooming relatives as to those of nongrooming relatives. How might we explain
kinship’s influence on these results?
9. In addition to kin selection and reciprocal altruism, a third mechanism leading to
nonrandom interaction of altruists has been suggested. Suppose altruists had an
easily detected phenotypic trait, perhaps a green beard. Then they could use the
following rule: “Do altruistic acts only for individuals who have green beards.”
Once the allele became common, most individuals carrying green beards would
not be related to one another, so this would not be a form of kin selection. How-
ever, there is a subtle flaw in this reasoning. Assuming that the genes controlling
beard color are at different genetic loci than the genes controlling altruistic behav-
ior, explain why green beards would not evolve.
10. Explain why punishment does not provide a ready solution to the problem of
cheating in mutualistic interactions.
KEY TERMS
Chapter 8
Big Brains and Long Lives
Life History Theory
Selective Pressures Favoring Large Brains
in Anthropoid Primates
What Do Monkeys Know about
One Another?
The Value of Studying Primate Behavior
L
arge brains and long life spans are two of the features that define the primate
order (Figure 8.1). Compared to most other animals, primates rely more heav-
ily on learning to acquire the knowledge and skills that they need to survive
and reproduce successfully. The complex behavioral strategies that we have explored
in the last few chapters depend on primates’ ability to respond flexibly in novel situ-
ations. Primates also have long periods of development and long life spans. Cogni-
tive complexity and longevity have become even more exaggerated among modern
humans, who live longer than any other primates and have relatively larger brains
198
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 199
than any other creatures on the planet. It is not a coincidence that primates
have both big brains and long lives; these traits are correlated across mam-
malian species. As we complete our discussion of the behavior and ecology
of contemporary primates and turn our attention to the history of the
human lineage, it is important to understand the forces that have shaped
the evolution of large brains and long life spans in the primate order.
Correlations tell us that two traits are related but not why this rela-
tionship exists. In this case, we have some reason to think that the causal
arrow goes from brain size to life span, not vice versa. We come to this con-
clusion because brains are expensive organs to maintain. Our brains
account for just 2% of our total body weight, but they consume about 20%
of our metabolic energy.
Natural selection does not maintain costly features like the brain unless
they confer important adaptive advantages. Moreover, the extent of investment
that organisms make in a particular feature will be linked to the benefit that
is derived from the investment. This is the same reason why you are usually
willing to spend more for something that you will use for a long time than for Primates are intelligent and
FIGURE 8.1
something you will use only once. Animals that live for a long time will derive long-lived. The first ape in
a greater benefit from the energy they expend on building and maintaining space was a four-year-old chimpanzee named Ham,
their brains than will animals that live for only a short time. who was trained to perform a variety of tasks as he
hurtled into space. In May 1961, three months
after Ham’s flight, Alan Shepard followed the chimp
into space. Ham was one of several dozen chim-
LIFE HISTORY THEORY panzees that NASA used to test the safety of
space travel for humans. Although Ham died at the
age of 27, a number of his fellow “astrochimps” are
Life history theory focuses on the evolutionary forces that shape still alive in their 40s. The survivors and some of
trade-offs between the quantity and quality of offspring and their descendants are now living in sanctuaries in
between current and future reproduction. New Mexico and southern Florida.
Birth and death mark the beginning and end of every individual’s life cycle. Between
these two end points, individuals grow, reach sexual maturity, and begin to reproduce.
Natural selection has generated considerable variation around this basic scheme. For
example, Pacific salmon are hatched in freshwater but spend their adult lives in the
open ocean. After years in the sea, they return to the streams where they were hatched
to lay or fertilize their own eggs; they die soon after they complete this journey. Opos-
sums, the only North American marsupial, produce their first litter at the age of one
year. Females have one to two litters per year, and live less than three years (Figure 8.2).
Lion females produce litters of up to six cubs at two-year intervals. Elephants conceive
for the first time at 10 years of age, have 22-month pregnancies, and give birth to sin-
gle infants at four- to nine-year intervals.
If natural selection favors increased reproductive success, why doesn’t it extend
the opossum’s life span, reduce the lion’s interbirth interval, or increase the elephant’s
litter size? The answer is that all organisms face fundamental trade-offs that constrain
their reproductive options. As we explained in Chapter 6, investment in one infant
limits investment in other offspring, so parents must make trade-offs between the
quality and quantity of offspring they produce. Organisms also face trade-offs
between current and future reproduction. All other things being equal, fast maturation
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 200
and early reproduction are advantageous because they increase the length
of the reproductive life span and reduce generation time. However, energy
devoted to current reproduction diverts energy from growth and mainte-
nance. If growth enhances reproductive success, then it may be advanta-
geous to grow large before beginning to reproduce. Thus, many kinds of
organisms have a juvenile phase in which they do not reproduce at all.
They do not become sexually mature until they reach a size at which the
payoffs of allocating energy to current reproduction exceed the payoffs of
continued growth. The same kind of argument applies to maintenance.
Energy that is diverted from current reproduction to maintenance enables
individuals to survive and reproduce successfully in the future.
Like humans, other primates age; as they get older, their physical abil-
ities deteriorate. They don’t run as fast, jump as high, or react as quickly
(Figure 8.3). Their teeth wear down, making it harder for them to chew
their food; and their joints deteriorate. Although humans are the only pri-
mates to experience menopause, the fertility of female primates of all
species declines when they reach old age. Males seem to reach peak phys-
ical condition in early adulthood and then decline.
The Virginia opossum, Didel-
FIGURE 8.2 At first glance, aging and death seem to be the inevitable effects of
phis virginiana, is the only
marsupial mammal in North America. Females pro- wear and tear on bodies. Organisms are complicated machines, like cars
duce many tiny fetuses, which make their way into or computers. A machine has many components that must function
the mother’s pouch, attach themselves to a nipple, together in order for it to work. It seems logical that the components in
and nurse for two to three months. Then they animals’ bodies simply wear out and break down, like a worn clutch or
emerge from the pouch and cling to their mother’s faulty hard disk. But this explanation of aging is flawed because the anal-
back. ogy between organisms and machines is not really apt. Every cell in an
organism contains all of the genetic information necessary to build a com-
plete, new body, and this genetic information can be used to repair damage. Wounds
heal and bones mend, and some organisms, such as frogs, can regenerate entire limbs.
Some organisms that reproduce asexually by budding or fission do not experience
senescence at all.
If senescence is not inevitable, why doesn’t natural selection do away with it? The
answer has to do with the relative magnitude of the benefits that animals can derive
from current reproduction or from living longer. Organisms could last longer if they
were built better. A Subaru costs much less to build than a Lexus, but it is also of lower
quality. As a consequence, a Lexus is not expected to break down as often as a Subaru
is. The same trade-off applies to organisms. Our teeth would last longer if they were
protected by a thicker covering of enamel, but building stronger teeth would require
more nutrients, particularly calcium. Building higher-quality organisms consumes time
and resources, thus reducing the organism’s growth rate and early fertility.
Senescence is at least partly the consequence of genes that increase fitness at early
ages and decrease it at later ages. Aging is favored by selection because traits that
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 201
increase fertility at young ages are favored at the expense of traits that increase
longevity.
The key to understanding this idea is to realize that selective pressures are much
weaker on traits that affect only the old. To see why, think about the fate of two
mutant alleles. One allele kills individuals before they reach adulthood, and the other
kills individuals late in their lives. Carriers of the allele that kills infants and juveniles
have a fitness of zero because none survive long enough to transmit the gene to their
descendants. Therefore, there will be strong selection against alleles with deleterious
effects on the young. In contrast, selection will have much less impact on a mutant
allele that kills animals late in their lives. Carriers of a gene that kills them late in life
will have already produced offspring before the effects of the gene are felt. Thus, a
mutation that affects the old will have limited effects on reproductive performance,
and there will be little or no selection against it. This means that genes with pleiotropic
effects that enhance early fertility but reduce fitness at later ages may be favored by
natural selection because they increase individual fitness.
The trade-offs between current and future reproduction and between FIGURE 8.3
Virtually all
the quantity and quality of offspring generate constellations of interrelated organisms experi-
ence senescence (aging). When this
traits. photograph was taken, this old male
chimpanzee, named Hugo, was missing
Animals that begin to reproduce early also tend to have small body sizes, small a lot of hair on his shoulders and back,
brains, short gestation times, large litters, high rates of mortality, and short life spans. he had lost a considerable amount of
Animals that begin to reproduce at later ages tend to have larger body sizes, larger weight, and his teeth had been worn
brains, longer gestation times, smaller litters, lower rates of mortality, and longer life down to the gums. Hugo died a few
spans (Figure 8.4). Life history traits are clustered together in this way because of the weeks later.
inherent trade-offs between current and future reproduction and between the quan-
tity and quality of offspring. Animals that begin to reproduce early divert energy from
growth and remain small. Animals that have small litters are able to invest more in
maintenance and extend their life spans. These clusters of correlated traits create a con-
tinuum of life history strategies that runs from fast to slow,
or from short to long. Opossums fall somewhere along the
fast/short end of the continuum; elephants lie at the slow/
long end.
shorter life history strategies. Other kinds of ecological factors may favor slower/
longer life histories. Suppose that there is severe competition for access to the
resources that animals need to reproduce successfully and that larger animals are
more successful in competitive encounters than small animals are. In this situation,
small animals will be at a competitive disadvantage, and it may be profitable to
invest more energy in growth, even if such an investment delays maturation. The life
history strategies that characterize organisms reflect the net effects of these kinds of
ecological pressures.
Primates fall toward the slow/long end of the life history continuum.
Most primatologists now believe that the enlargement and reorganization of the brain
in anthropoids was linked to the competitive pressures produced by sociality. In social
groups, animals compete for food, mates, grooming part-
ners, and other valuable resources. They also form social
bonds that influence their participation in coalitions,
exchange networks, access to resources, and so on (Figure
8.6). The larger a group becomes, the more difficult it gets
to sustain social bonds and keep track of relationships
within the group. The ability to operate effectively in this
complicated social world may reward greater flexibility in
behavior and favor expansion of the parts of the brain
that are linked to learning and planning. This idea is
called the social intelligence hypothesis.
An alternative set of hypotheses links increased brain
size to ecological challenges. Monkeys feed mainly on
plants and include many different plant species in their
diets. They must evaluate the ripeness, nutritional con-
tent, and toxicity of their food items. Katharine Milton of Many primates live in complex social groups.
FIGURE 8.6
the University of California, Berkeley has argued that this Geladas form one-male units, which aggregate
together to form large bands composed of hundreds of individuals.
task may be particularly taxing for frugivores, whose
foods are patchily distributed in time and space. Success-
ful exploitation of this frugivorous feeding niche may favor better memory, greater
capacities for learning, and the ability to plan a course of action.
Katherine Gibson of the University of Texas and Sue Parker of Sonoma State Uni-
versity offer a second ecological hypothesis. They believe that natural selection has
favored enhanced cognitive capacities among primates because primates rely heavily
on extracted foods that require complex processing techniques. For example, some pri-
mates eat hard-shelled nuts that must be cracked open with stones or smashed against
a tree trunk. Others dig up roots and tubers; extract insect larvae from bark, wood pith,
or dung; and crack open pods to obtain seeds (Figure 8.7). Extracted foods are valu-
able elements in primate diets because they tend to be rich sources of protein and
energy. However, they require complicated, carefully coordinated techniques to process.
A third model, developed by Simon Reader of Utrecht University and Kevin
Laland of the University of St. Andrews, proposes that natural selection has favored
changes in the primate brain that facilitate behavioral flexibility. In particular, Reader
and Laland emphasize the ability to invent appropriate solutions to novel problems
(Figure 8.8) and the ability to learn new behaviors from conspecifics. The benefits
derived from innovation and social learning generated selective pressures that favored
expansion and development of the parts of the brain linked to learning, planning, and
behavioral flexibility. Reader and Laland emphasize that the ability to innovate and
learn from others might enhance animals’ ability to cope with both ecological and
social challenges.
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 204
Primates sometimes
FIGURE 8.7
exploit foods that
are difficult to extract. Here, (a) a male
chimpanzee pokes a long twig into a hole in
a termite mound and extracts termites, and
(b) a capuchin monkey punctures an
eggshell and extracts the contents. ((b)
Photograph courtesy of Susan Perry.)
(a) (b)
Cerebellum Occipital
lobe
Frontal
Brain stem Temporal lobe Sylvian fissure lobe
Brain stem
Parietal Temporal lobe
lobe (c) Human
Frontal lobe
The brains of a large prosimian, the indri (a); a macaque
FIGURE 8.9
(b); and a human (c). There are three main components
Occipital
of the brain: the hindbrain, which contains the cerebellum and medulla (part of
lobe
the brain stem); the midbrain, which contains the optic lobe (not visible here);
and the forebrain, which contains the cerebrum. The cerebrum is divided into
four main lobes: occipital, parietal, frontal, and temporal. The forebrain is
Cerebellum Sylvian fissure greatly expanded in primates and other mammals, and much of the gray matter
(which is made up of cell bodies and synapses) is located on the outside of the
Brain stem Temporal
cerebrum in a layer called the “cerebral cortex.” The neocortex is a component
lobe
of the cerebral cortex, and in mammals the neocortex covers the surface of vir-
(b) Macaque tually the entire forebrain.
Finally, there is some evidence that behavioral flexibility and social learning are
linked to brain evolution. Reader and Laland surveyed the primate literature for infor-
mation about three measures of behavioral flexibility: (1) reports of behavioral inno-
vation, which they defined as novel solutions to ecological or social problems; (2)
examples of social learning, the acquisition of skills and information from others; and
(3) observations of tool use. They demonstrated that these measures of behavioral flex-
ibility are closely linked to the executive brain ratio (the size of the executive brain in
relation to the brain stem). Primates with relatively large executive brains are more
likely to innovate, learn from others, and use tools than primates with relatively small
executive brains are. In addition, primates seem to be more flexible in their foraging
behavior than in their social behavior. In the long list of examples of innovations and
socially learned behaviors that Reader and Laland compiled, foraging innovations pre-
dominate. In these comparative analyses, there is no consistent relationship between
social learning and group size.
Great apes do not fit the social intelligence hypothesis very well.
Great apes have larger brains in relation to their body size than monkeys do, but
they live in smaller groups than many monkeys do. Although chimpanzees and bono-
bos may live in communities that include as many as 50 individuals, gorillas live in
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 206
(a) (b)
Chimpanzees use tools to help them obtain certain kinds of foods. (a) This
FIGURE 8.10
chimpanzee at Gombe Stream National Park in Tanzania is using a long stick to
dip for ants. (b) Here a chimpanzee uses a large stone to hammer open hard-shelled nuts.
much smaller groups, and orangutans are largely solitary. This obviously poses a prob-
lem for the social intelligence hypothesis.
Richard Byrne points out that great apes make use of more complicated foraging
techniques than other primates do, enabling them to feed on some foods that other pri-
mates cannot process. For example, virtually all plant foods that mountain gorillas rely
on are well defended by spines, hard shells, hooks, and stingers. Each of their food
items requires a particular routine—a complicated sequence of steps structured in a par-
ticular way. Many of the foods that orangutans feed upon are also difficult to process.
Great apes sometimes use tools to obtain access to certain foods that are not oth-
erwise available to them. Chimpanzees poke twigs into holes of termite mounds and
anthills, use leaves as sponges to mop up water from deep holes, and employ stones
as hammers to break open hard-shelled nuts (Figure 8.10). Recently, Carel van Schaik
and his colleagues observed Sumatran orangutans using sticks to probe for insects and
to pry seeds out of the husks of fruit. Although bonobos and gorillas have not been
observed using tools in the wild, all great apes are adept tool users in captive settings.
Living in com-
FIGURE 8.12
plex social
groups may have been the selective fac-
tor favoring large brains and intelligence
among primates.
off steam or venting their aggression, they would choose a target at random. Thus, the
monkeys seem to know that certain individuals are somehow related.
Because kinship and dominance rank are major organizing principles in most
primate groups, it makes sense to ask whether monkeys also understand third-party
rank relationships. The most direct evidence that monkeys understand third-party rank
relationships comes from two playback experiments conducted on a group of baboons
in the Okavango Delta of Botswana who have been studied for the last 15 years by
Seyfarth, Cheney, and their colleagues. In this group, dominance relationships were
stable, and females never responded submissively toward lower-ranking females.
In one experiment that Seyfarth and Cheney designed, females listened to a record-
ing of a female’s grunt followed by another female’s submissive fear barks. Female
baboons responded more strongly when they heard a higher-ranking female respond-
ing submissively to a lower-ranking female’s grunt than when they heard a lower-
ranking female responding submissively to a higher-ranking female’s grunt. Thus,
females were more attentive when they heard a sequence of calls that did not corre-
spond to their knowledge of dominance rank relationships among other females. Con-
trol experiments excluded the possibility that females were reacting simply to the fact
that they had not heard a particular sequence of calls before. The pattern of responses
suggests that females knew the relative ranks of other females in their group and were
particularly interested in the anomalous sequence of calls.
In a second experiment, Thore Bergman and Jacinta Beehner collaborated with
Seyfath and Cheney to probe the baboons’ knowledge of the hierarchical nature of
rank relationships in groups with matrilineal dominance ranks. Using the same basic
experimental paradigm, researchers played sequences of vocalizations that simulated
rank reversals within lineages and rank reversals between lineages. As young female
baboons mature, they often rise in rank above their older sisters and other female kin.
Thus, changes in the relative rank of females in the same lineage are part of the nor-
mal course of rank acquisition. However, changes in the relative ranks of unrelated
females are much less common. The baboons reacted much more strongly to simulated
rank reversals between lineages than to simulated rank reversals within lineages.
Again, the researchers were careful to control for confounding variables, such as rank
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 209
Ally
Aggressor
Victim Complex fitness calculations may be involved
FIGURE 8.13
in decisions about whether to join a coalition.
By helping the victim against the aggressor, the ally increases the fitness
of the victim and decreases the fitness of the aggressor.
distance and novelty. This result suggests that the females understood the relative ranks
of other females and that they understood that changes in rank relationships within
lineages are not the same as changes in rank relationships between lineages.
Even the simplest coalition is a complex interaction. When coalitions are formed,
at least three individuals are involved, and several different kinds of interactions are
going on simultaneously (Figure 8.13). Consider the case in which one monkey (the
aggressor) attacks another monkey (the victim). The victim then solicits support from
a third party (the ally), and the ally intervenes on behalf of the victim against the
aggressor. The ally behaves altruistically toward the victim, giving support to the vic-
tim at some potential cost to itself. At the same time, however, the ally behaves aggres-
sively toward the aggressor, imposing harm or energy costs on the aggressor. Thus, the
ally simultaneously has a positive effect on the victim and a negative effect on the
aggressor. Under these circumstances, decisions about whether or not to intervene in
a particular dispute may be quite complicated. Consider a female who witnesses a dis-
pute between two of her offspring. Should she intervene? If so, which of her offspring
should she support? When a male bonnet macaque is solicited by a higher-ranking
male against a male who frequently supports him, how should he respond? In each
case, the ally must balance the benefits to the victim, the costs to the opponent, and
the costs to itself (Figure 8.14).
Given the complexity of even simple coalitions, knowledge of third-party rela-
tionships may be valuable because it enables individuals to predict how others will
behave. Thus, animals who understand the nature of third-party relationships may
have a good idea about who will support them and who will intervene against them
in confrontations with particular opponents, and they may also be able to tell which
of their potential allies are likely to be most effective in coalitions against their
opponents.
Monkeys seem to make use of information about third-party relationships when
they recruit support in coalitions. Male bonnet macaques are most likely to intervene
Primates form
FIGURE 8.14
coalitions that
are more complicated than the coalitions
of most other animals. (a) In a captive
bonnet macaque group, members of
opposing factions confront one another.
(b) Two capuchins jointly threaten a
third individual who is not visible in the
picture. ((b) Photograph courtesy of
Susan Perry.)
(a) (b)
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 210
in ongoing conflicts if they outrank both of the participants in the original conflict.
Males seem to make use of this information when they recruit support from potential
allies; they selectively solicit males who outrank both themselves and their opponents.
Their ability to do this consistently relies on their knowledge of the rank relationships
among other males.
Susan Perry and her colleagues at the University of California, Los Angeles have
also studied monkeys’ use of third-party information in coalitions. Capuchins follow
three basic rules when they form coalitions: (1) support females against males, (2) sup-
port dominants against subordinates; and (3) support close associates against others.
If capuchins “understand” these rules, particularly the last one, they are not expected
to recruit support against dominant opponents or from males who have closer rela-
tionships with their opponents than with themselves. This is exactly what they do.
Their ability to adhere to these rules suggests that they understand the nature of rela-
tionships among other group members.
We are beginning to gain insight into what monkeys and apes know about
others’ minds.
Monkeys and apes seem to be very good at predicting what other animals will do
in particular situations. For example, we have seen that vervets groom monkeys who
support them in coalitions, female langurs with newborn infants are fearful of new res-
ident males, baboons express surprise when low-ranking animals elicit signs of sub-
mission from higher-ranking animals, and capuchins don’t try to recruit support from
monkeys who have closer bonds to their opponents than to themselves. These exam-
ples indicate that monkeys can predict what others will do and adjust their behavior
accordingly.
Monkeys’ ability to predict what other individuals will do may not seem very
remarkable. After all, we know that many animals are very good at learning to make
associations between one event and another. In the laboratory, rats, pigeons, monkeys,
and many other animals can learn to pull a lever, push a button, or peck a key to
obtain food. These are examples of associative learning, the ability to track contin-
gencies between one event and another. Monkeys’ ability to predict what others will
do in particular situations might be based on sophisticated associative learning capac-
ities, prodigious memory of past events, and perhaps some understanding of concep-
tual categories like kinship and dominance. On the other hand, it is also possible that
monkeys’ ability to predict what others will do is based on their knowledge of the men-
tal states of others—what psychologists call a theory of mind.
It may seem relatively unimportant whether monkeys and apes rely on associative
learning to predict what others will do or whether they have a well-developed theory
of mind. However, there may be some things that animals cannot do unless they under-
stand what is going on in other animals’ minds. For example, some researchers think
that effective deception requires the ability to manipulate or take advantage of others’
beliefs about the world. In the 1960s, Emil Menzel, formerly of the State University
of New York at Stony Brook, conducted a landmark set of experiments about chim-
panzees’ ability to find and communicate about the location of hidden objects (Figure
8.15). In one set of experiments, Menzel showed one chimpanzee where a food item
was hidden, then released the knowledgeable chimpanzee and his companions into
their enclosure. The group quickly learned to follow their knowledgeable companion;
but he just as quickly learned that when he led others to the hidden food, he would
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 211
not get a very big share of it. Menzel noticed that the knowledgeable chim-
panzee sometimes led his companions in the wrong direction and then
dashed off and grabbed the hidden treasure. How did the knowledgeable
chimpanzee work out this tactic? He might have understood that his
knowledge differed from the knowledge of other group members and then
have come up with a way to take advantage of this discrepancy effectively.
If this is what he did, then we would conclude that he had a well-developed
theory of mind.
Although some researchers, such as Richard Byrne, suggest that decep-
tion does not rely on theory of mind, it seems clear that a theory of mind
would allow for more complicated and successful deceptions. Similarly, the
ability to pretend, empathize, take another’s perspective, read minds, con-
sole, imitate, and teach relies on knowing what others know or how they
feel. Humans do all of these things, but it is not clear whether other pri-
mates do.
Subordinate
Anthropology in Leipzig, Germany, reasoned that the chimpanzees might have per-
formed poorly in Povinelli’s experiments because the situation required them to rea-
son about cooperative interactions involving food, and these kinds of interactions are
rare among chimpanees. So, they developed a protocol that evaluates indviduals’ abil-
ity to take advantage of discrepancies between their own knowledge and other indi-
viduals’ knowledge in a competitive situation. In their experiments, they paired
subordinate and dominant chimpanzees in the configuration illustrated in Figure 8.17.
These experiments take advantage of the fact that subordinates cannot obtain food
when dominants are present. Here, two pieces of food are visible to the subordinate,
but the dominant can see only one; the other is hidden behind a barrier. Hare and his
colleagues predicted that if the subordinate knew what the dominant could see, the
subordinate would head for the piece of food that the dominant could not see, hop-
ing to consume the hidden food item while the dominant was occupied with the other
piece. This is exactly what the subordinate chimpanzees did. So, the chimpanzees
seemed to understand what other chimpanzees know.
Laurie Santos of Yale University and her colleagues applied the same reasoning in
designing experiments with free-ranging rhesus macaques on Cayo Santiago. In these
experiments, monkeys were given the chance to “steal” food from two human exper-
imenters. In one experiment, one experimenter was facing toward the monkey, and the
other was facing away. The monkeys were more likely to approach the experimenter
who was facing away than the experimenter who was facing forward. In other exper-
iments, the monkeys selectively approached experimenters whose faces were pointed
away and experimenters whose eyes were averted (Figure 8.18). In another set of
experiments, Santos placed food rewards in two transparent boxes covered with bells.
In one box, the ringers on the bells were removed; creating one noisy box and one
quiet box. The experimenter baited the boxes, shook the boxes to display their audi-
tory qualities, and then walked away and hid his face. The monkeys showed stong
preferences for the quiet container, suggesting that they knew what the experimenter
could hear. Santos and her colleagues argue that rhesus monkeys have accurate per-
ceptions about what others see and hear in these competitive situations.
Although apes and monkeys are able to solve many complex cognitive tasks, there
are still substantial differences between the cognitive skills of humans and other pri-
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 213
Cameraperson 2 Cameraperson 2
Competitor 1 Competitor 2
Competitor 1 Competitor 2
Monkey Monkey
Cameraperson 1 Cameraperson 1
Number of subjects
who had food that they could potentially “steal.” The graphs show
16 p < 0.0001 16 p < 0.0001
14 14 that monkeys were much more likely to approach the person who
12 12 was looking away from them than the person who was facing
10 10 them, and they were more likely to approach the person who was
8 8
6 6 looking toward the food than the person who was looking away
4 4 from the food. This suggests that monkeys know what others can
2 2 and cannot see. (From Figure 1, in J. I. Flombaum and L. R. San-
0 0
Competitor 1 Competitor 2 Competitor 1 Competitor 2 tos, 2005, “Rhesus Monkeys Attribute Perceptions to Others,”
(a) Experiment 1 (b) Experiment 2 Current Biology 15:447–452.)
mates. These differences are most pronounced in tasks that involve social learning,
communication, and knowledge of others’ minds. In a comprehensive comparative
study of ape cognition, Esther Herrmann and her colleagues from the Max Planck
Institute of Evolutionary Anthropology evaluated the performance of 105 two-year-
old humans on a battery of cognitive tasks and compared this with the performance
of 106 chimpanzees and 32 orangutans of all ages on the same tasks. Some tasks
focused on cognition about the physical world, such as tracking a reward after it has
been moved or using a tool to retrieve a reward that is out of reach. Other tasks
focused on social cognition, such as solving a novel problem after observing the
demonstration of a solution or following a gaze to a target. There was little difference
in the performance of children and apes on tasks that involved physical cognition (Fig-
ure 8.19). However, there were much greater differences in the social domain. The
human children were significantly more successful than the chimpanzees and orang-
utans on tasks that required social learning, communication, and knowledge of oth-
ers’ minds. Herrmann and her colleagues hypothesize that humans have “evolved some
specialized socio-cognitive skills (beyond those of primates in general) for living and
exchanging knowledge in cultural groups: communicating with others, learning from
others, and ‘reading the mind’ of others in especially complex ways.” We will come
back to this idea in Part Four, when we discuss the evolution of the human capacity
for culture.
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 214
Children and apes were presented with an identical set of tasks that examined
FIGURE 8.19
their physical and social cognition. Children and apes showed no differences in
tasks based on physical cognition. But children were considerably more successful than apes on tasks
that were based on social cognition. (From Figure 1, in E. Hermann, J. Call, M. V. Hernández-Lloreda,
B. Hare, and M. Tomasello, 2007, “Humans Have Evolved Specialized Skills of Social Cognition: The
Cultural Intelligence Hypothesis.” Science 317:1360–1366.)
FURTHER READING
Kaplan, H., T. Mueller, S. Gangestad, and J. B. Lancaster. 2003. Neural capital and
life span evolution among primates and humans. In The Brain and Longevity, ed.
C. E. Finch, J.-M. Robine, and Y. Christen. Berlin: Springer. Pp. 69–97.
Reader, S. M., and K. N. Laland. 2002. Social intelligence, innovation, and enhanced
brain size in primates. Proceedings of the National Academy of Sciences of the
United States of America 99:4436–4441.
Whiten, A. W., and R. W. Byrne, eds. 1997. Machiavellian Intelligence II: Extensions
and Evaluations. New York: Cambridge University Press.
1. There is a positive correlation between brain size and longevity in animal species.
One interpretation of this correlation is that selection for longer life spans was the
primary force driving the evolution of large brains. An alternative interpretation
is that selection for larger brains was the primary force driving the evolution of
longer life spans. Explain which of these interpretations is more likely to be cor-
rect, and why this is the case.
2. We have argued that natural selection is a powerful engine for generating adap-
tations. If that is the case, then why do organisms grow old and die? Why can’t
natural selection design an organism that lives forever?
3. Life history traits tend to be bundled together in particular ways. Explain how
these traits are combined and why we see these kinds of combinations in nature.
4. Primates evolved from small-bodied insectivores that were arrayed somewhere
along the fast/short end of the life history continuum. What ecological factors are
thought to have favored the shifts toward slower/longer life histories in early pri-
mates, monkeys, and apes?
5. Primates take a relatively long time to grow up, compared to other animals. Con-
sider the costs and benefits of this life history pattern from the point of view of
the growing primate and its mother.
6. What do comparative studies of the size and organization of primate brains tell
us about the selective factors that shaped the evolution of primate brains? What
are the shortcomings of these kinds of analyses?
7. Monkeys are quite skilled in navigating complicated social situations that they
encounter in their everyday lives. They seem to know what others will do in par-
ticular situations and are able to respond appropriately. However, monkeys con-
sistently fail theory-of-mind tests in the laboratory. How can we reconcile these
two observations?
8. Monkeys seem to have some concept of kinship. What evidence supports this
idea? What kind of variation might you expect to find in monkeys’ concepts of
kinship within and between species?
9. Suppose that you were studying a group of monkeys and you discovered con-
vincing evidence of empathy or deception. How and why would these data sur-
prise your colleagues?
08_Primate Life Histories [4p]:How Humans Evolved [5e] 10/10/08 11:57 AM Page 216
KEY TERMS
P a rt T h re e
Chapter 9
Continental Drift and Climate Change
The Methods of Paleontology
The Evolution of the Early Primates
The First Anthropoids
The Emergence of the Hominoids
D
uring the Permian and early Triassic periods (Table 9.1), much of the
world’s fauna was dominated by therapsids, a diverse group of reptiles that
possessed traits, such as being warm-blooded and covered with hair (Fig-
ure 9.1), that linked them to the mammals that evolved later. At the end of the Tri-
assic, most therapsid groups disappeared, and dinosaurs radiated to fill all of the
niches for large, terrestrial animals. One therapsid lineage, however, evolved and
diversified to become the first true mammals. These early mammals were probably
mouse-sized, nocturnal creatures that fed mainly on seeds and insects. They had inter-
nal fertilization but still laid eggs. By the end of the Mesozoic era, 65 million years
218
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 219
TABLE 9.1
The geological timescale.
Period
Begins
Era Period Epoch (mya) Notable Events
ago (mya), placental and marsupial mammals that bore live young had evolved. With
the extinction of the dinosaurs at the beginning of the next era (the Cenozoic) came
the spectacular radiation of the mammals. All of the modern descendants of this
radiation—including horses, bats, whales, elephants, lions, and primates—evolved
from creatures something like a contemporary shrew (Figure 9.2).
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 220
(a)
animals something like present-day apes. In later chapters, we recount the transfor-
mation from hominoid to hominin. We will introduce the first members of the human
tribe, Hominini; and then the first members of our own genus, Homo; and finally the
first known representatives of our own species, Homo sapiens. We know something
about each step in this process, although far more is known about recent periods than
about periods in the most distant past. We will see, however, that there is still a great
deal left to be discovered and understood.
When we think about the evolution of modern humans, we usually picture early
humans wandering over open grasslands dotted with acacia trees—the same breath-
taking scenery that we see in wildlife documentaries. As we shift the time frame for-
ward through millions of years, the creatures are altered but the backdrop is
unchanged. Our mental image of this process is misleading, however, because the
scenery has changed along with the cast of characters (Figure 9.3).
It is important to keep this fact in mind because it alters our interpretation of the
fossil record. Remember that evolution produces adaptation, but what is adaptive in
one environment may not be adaptive in another environment. If the environment
remained the same over the course of human evolution, then the kinds of evolution-
ary changes observed in the hominin fossil record (such as increases in brain size,
bipedalism, and prolonged juvenile dependence) would have to be seen as steady
improvements in the perfection of human adaptations: evolution would progress
toward a fixed goal. But if the environment varied through time, then evolution would
have to track a moving target. In this scenario, new characteristics seen in fossils would
not have to represent progress in a single direction. Instead, these changes might have
been adaptations to changing environmental condi-
tions. As we will see, the world has become much
colder and drier in the last 20 million years and par-
ticularly variable in the last 800,000 years, and these
changes likely altered the course of human evolution.
If the world had become warmer rather than colder
during this period, then our human ancestors would
probably have remained in the safety of the trees and
would not have become terrestrial or bipedal. We
would probably be stellar rock climbers but poor
marathon runners.
LAURASIA
EA
A
G
N
120° 80° 40° 80° 120° 120° 80° 80° 120°
PA
GON
DW
AN
AL
AN
D
er ic a
Am
rth Eurasia
o
N
Ind
ut ca
h
Mada-
gascar
a
trali
Aus
Antarctica
65 mya Present
The arrangement of the continents has changed considerably over the last 180
FIGURE 9.4
million years.
continental drift. The continents are not fixed in place; instead, the enormous, rela-
tively light plates of rock that make up the continents slowly wander around the
globe, floating on the denser rock that forms the floor of the deep ocean. About 200
mya, all of the land making up the present-day continents was joined together in a
single, huge landmass called Pangaea. About 125 mya, Pangaea began to break apart
into separate pieces (Figure 9.4). The northern half, called Laurasia, included what
is now North America and Eurasia minus India; the southern half, Gondwanaland,
consisted of the rest. By the time the dinosaurs became extinct 65 mya, Gondwana-
land had broken up into several smaller pieces. Africa and India separated, and India
headed north, eventually crashing into Eurasia, while the remainder of Gondwana-
land stayed in the south. Eventually, Gondwanaland separated into South America,
Antarctica, and Australia, and these continents remained isolated from each other for
many millions of years. South America did not drift north to join North America until
about 5 mya.
Continental drift is important to the history of the human lineage for two reasons:
First, oceans serve as barriers that isolate certain species from others, so the position
of the continents plays an important role in the evolution of species. As we will see,
the long isolation of South America creates one of the biggest puzzles in our knowl-
edge of primate evolution. Second, continental drift is one of the engines of climate
change, and climate change has fundamentally influenced human evolution.
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 223
15
0
10 The gray points
FIGURE 9.5
are estimates of
1 average world temperature based on the
Temperature (C°)
δ 18O (0/00)
2 cores. The red line plots a statistically
“smoothed” average value. The variabil-
0 ity around the line represents both
3
measurement error and rapid tempera-
ture fluctuations that last less than a
4
–5 million years. As we will see (Figure
Paleocene Eocene Oligocene Miocene Pliocene 12.1), most variability in the last sev-
5 eral million years has been the result of
70 60 50 40 30 20 10 0 fluctations in world temperature, some
Time (millions of years) lasting only a few centuries.
The climate has changed substantially during the last 65 million years—first
becoming warmer and less variable, then cooling, and finally fluctuating widely
in temperature.
The size and orientation of the continents have important effects on climate. Very
large continents tend to have severe weather. This is why Chicago has much colder
winters than London, even though London is much farther north. Pangaea was much
larger than Asia and is likely to have had very cold weather in winter. When continents
restrict the circulation of water from the tropics to the poles, world climates seem to
become cooler. These changes, along with other, poorly understood factors, have led
to substantial climate change. Figure 9.5 summarizes changes in global temperature
during the Cenozoic era. Box 9.1 explains how climatologists reconstruct ancient
climates.
To give you some idea of what these changes in temperature mean, consider that
during the period of peak warmth in the early Miocene, palm trees grew as far north
as what is now Alaska, rich temperate forests (like those in the eastern United States
today) extended as far north as Oslo (Norway), and only the tallest peaks in Antarc-
tica were glaciated.
In certain kinds of geological settings, the bones of dead organisms may be preserved
long enough for the organic material in the bones to be replaced by minerals (miner-
alized) from the surrounding rock. Such natural copies of bones are called fossils. Sci-
entists who recover, describe, and interpret fossil remains are called paleontologists.
A great deal of what we know about the history of the human lineage comes from
the study of fossils. Careful study of the shapes of different bones tells us what early
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 224
Box 9.1
Using Deep-Sea Cores
to Reconstruct Ancient
Climates
Beginning about 35 years ago, oceanographers began a below the surface). Data from these cores have allowed sci-
program of extracting long cores from the sediments that lie entists to make much more detailed and accurate recon-
on the floor of the deep sea (about 6,000 m, or 20,000 feet, structions of ancient climates. Figure 9.5 shows the ratio of
Evaporation:
more 16O
Runoff:
16O 18O more 16O
16O 18O
The ocean:
baseline 16O
16O 18O
(a) World is warm
Evaporation:
more 16O
Glacial ice:
16O 18O more 16O
Because water
FIGURE 9.6
evaporating from
the sea is enriched in 16O, so is precipi-
16O 18O
tation. (a) When the world is warm, this
water returns rapidly to the sea, and the
concentration of 16O in seawater is The ocean:
unchanged. (b) When the world is cold, less 16O
much precipitation remains on land as
glacial ice, and so the sea becomes
depleted in 16O. 16O 18O
(b) World is cold
224
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 225
two isotopes of oxygen, 16O:18O, derived from different lay- warm enough that few glaciers form at high latitudes, the
ers of a deep-sea cores. Because different layers of the cores precipitation that falls on the land returns to the sea, and the
were deposited at different times over the last 65 million ratio of the two isotopes of oxygen remains unchanged (Fig-
years and have remained nearly undisturbed ever since, ure 9.6a). When the world is colder, however, much of the
they give us a snapshot of the relative amounts of 16O and snow falling at high latitudes is stored in immense conti-
18
O in the sea when the layer was deposited on the ocean nental glaciers like those now covering Antarctica (Figure
floor. 9.6b). Because the water locked in glaciers contains more
These ratios allow us to estimate ocean temperatures in 16
O than the ocean does, the proportion of 18O in the ocean
the past. Water molecules containing the lighter isotope of becomes greater. Therefore, the concentration of 18O in sea-
oxygen, 16O, evaporate more readily than molecules con- water increases when the world is cold and decreases when
taining the heavier isotope, 18O, do. Snow and rain have a it is warm. This means that scientists can estimate the tem-
higher concentration of 16O than the sea does because the perature of the oceans in the past by measuring the ratio of
water in clouds evaporates from the sea. When the world is 16
O to 18O in different layers of deep-sea cores.
hominins were like—how big they were, what they ate, where they lived, how they
moved, and even something about how they lived. When the methods of systematics
described in Chapter 4 are applied to these materials, they also can tell us something
about the phylogenetic history of long-extinct creatures. The kinds of plant and ani-
mal fossils found in association with the fossils of our ancestors tell us what the envi-
ronment was like—whether it was forested or open, how much it rained, and whether
rainfall was seasonal.
There are several radiometric methods for estimating the age of fossils.
1. Potassium–argon dating is used to date the age of volcanic rocks found in asso-
ciation with fossil material. Molten rock emerges from a volcano at a very high
temperature. As a result, all of the argon gas is boiled out of the rock. After this,
any argon present in the rock must be due to the decay of potassium. Because
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 226
this occurs at a known and constant rate, the ratio of potassium to argon can
be used to date volcanic rock. Then, if a fossil is discovered in a geological stra-
tum (“layer”; plural strata) lying under the stratum that contains the volcanic
rock, paleontologists can be confident that the fossil is older than the rock. A
new variant of this technique, called argon–argon dating because the potassium
in the sample is converted to an isotope of argon before it is measured, allows
more accurate dating of single rock crystals.
2. Carbon-14 dating (or radiocarbon dating) is based on an unstable isotope of car-
bon that living animals and plants incorporate into their cells. As long as the
organism is alive, the ratio of the unstable isotope (carbon-14) to the stable iso-
tope (carbon-12) is the same as the ratio of the two isotopes in the atmosphere.
Once the animal dies, carbon-14 starts to decay into carbon-12 at a constant
rate. By measuring the ratio of carbon-12 to carbon-14, paleontologists can esti-
mate the amount of time that has passed since the organism died.
3. Thermoluminescence dating is based on an effect of high-energy nuclear parti-
cles traveling through rock. These particles come from the decay of radioactive
material in and around the rock and from cosmic rays that bombard the Earth
from outer space. When they pass through rock, these particles dislodge elec-
trons from atoms, so the electrons become trapped elsewhere in the rock’s crys-
tal lattice. Heating a rock relaxes the bonds holding the atoms in the crystal
lattice together. All of the trapped electrons are then recaptured by their respec-
tive atoms—a process that gives off light. Researchers often find flints at archae-
ological sites that were burnt in ancient campfires. It is possible to estimate the
number of trapped electrons in these flints by heating them in the laboratory and
measuring the amount of light given off. If the density of high-energy particles
currently flowing through the site is also known, scientists can estimate the
length of time that has elapsed since the flint was burned.
4. Electron-spin-resonance dating is used to determine the age of apatite crystals,
an inorganic component of tooth enamel, according to the presence of trapped
electrons. Apatite crystals form as teeth grow, and initially they contain no
trapped electrons. These crystals are preserved in fossil teeth and, like the burnt
flints, are bombarded by a flow of high-energy particles that generate trapped
electrons in the crystal lattice. Scientists estimate the number of trapped electrons
by subjecting the teeth to a variable magnetic field—a technique called “electron
spin resonance.” To estimate the number of years since the tooth was formed,
paleontologists must once again measure the flow of radiation at the site where
the tooth was found.
Different radiometric techniques are used for different time periods. Methods
based on isotopes that decay very slowly, such as potassium-40, work well for fossils
from the distant past. However, they are not useful for more recent fossils, because
their “clock” doesn’t run fast enough. When slow clocks are used to date recent events,
large errors can result. For this reason, potassium–argon dating usually cannot be used
to date samples less than about 500,000 years old. Conversely, isotopes that decay
quickly, such as carbon-14, are useful only for recent periods, because all of the unsta-
ble isotopes decay in a relatively short period of time. Thus, carbon-14 can be used
only to date sites that are less than about 40,000 years old. The development of ther-
moluminescence dating and electron-spin-resonance dating is important because these
methods allow us to date sites that are too old for carbon-14 dating but too young for
potassium–argon dating.
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 227
Radiometric dating methods are problematic for two reasons. First, a particular
site may not always contain material that is appropriate for radiometric dating. Sec-
ond, radiometric methods have relatively large margins for error. These drawbacks
have led scientists to supplement these absolute methods with other, relative methods
for dating fossil sites.
One such relative method is based on the remarkable fact that, every once in a
while, the Earth’s magnetic field reverses itself. This means, for example, that com-
passes now pointing north would at various times in the past have pointed south (if
compasses had been around then, that is). The pattern of magnetic reversals is not the
same throughout time, so for any given time period the pattern is unique. But the pat-
tern for a given time is the same throughout the world. We know what the pattern is
because when certain rocks are formed, they record the direction of the Earth’s mag-
netic field at that time. Thus, by matching up the pattern of magnetic reversals at a par-
ticular site with the well-dated sequence of reversals from the rest of the world,
scientists are able to date sites.
Another approach is to make use of the fact that sometimes the fossils of interest
are found in association with fossils of other organisms that existed for only a limited
period of time. For example, during the last 20 million years or so there has been a
sequence of distinct pig species in East Africa. Each pig species lived for a known
period of time (according to securely dated sites). This means that some East African
materials can be accurately dated from their association with fossilized pig teeth.
During the first two-thirds of the Mesozoic, the forests of the world were dominated
by the gymnosperms, trees like contemporary redwood, pine, and fir. With the
breakup of Pangaea during the Cretaceous, a revolution in the plant world occurred.
Flowering plants, called angiosperms, appeared and spread. The evolution of the
angiosperms created a new set of ecological niches for animals. Many angiosperms
depend on animals to pollinate them, and they produce showy flowers with sugary
nectar to attract pollinators. Some angiosperms also entice animals to disperse their
seeds by providing nutritious and easily digestible fruits. Arboreal animals that could
find, manipulate, chew, and digest these fruits could exploit these new niches. Primates
were one of the taxonomic groups that evolved to take advantage of these opportu-
nities. Tropical birds, bats, insects, and some small rodentlike animals probably com-
peted with early primates for the bounty of the angiosperms.
To understand the evolutionary forces that shaped the early radiation of the
primates, we need to consider two related questions. First, what kind of animal did
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 228
Plesiadapiforms Prosimians
Plesiadapi- Claws
FIGURE 9.7
forms were
once thought to be primates, but
Auditory Large incisors Auditory
closer analysis indicates they lack No opposable
bulla bulla
many of the defining characteristics big toe
of early modern primates. For exam- Small incisors,
ple, they had claws instead of nails; Gap large canine
their eye orbits were not fully Plesiadapis (diastema)
encased in bone; their eyes were
placed on the sides of the head, so or
the fields of vision of the two eyes Large incisors,
small canine
did not converge; and in some
species the big toe was not
opposable.
natural selection have to work with? Second, what were the selective pressures that
favored this suite of traits in ancient primates?
The plesiadapiforms, a group of fossil animals found in what is now Montana,
Colorado, New Mexico, and Wyoming, give us some clue about what the earliest pri-
mates were like. Plesiadapiforms are found at sites that date from the Paleocene epoch,
65 to 54 mya, a time so warm and wet that broadleaf, evergreen forests extended to
60°N (latitude 60° north, near present-day Anchorage in Alaska). The plesiadapiforms
varied from tiny, shrew-sized creatures to animals as big as a marmot (Figure 9.7). It
seems likely that they were solitary quadrupeds with a well-developed sense of smell.
The teeth of these animals are quite variable, suggesting that they had a wide range
of dietary specializations. Some members of this group were probably terrestial, some
were arboreal quadrupeds, and others may have been adapted for gliding. Most of the
plesiadapiforms had claws on their hands and feet, and they did not have binocular
vision.
The recent discovery of a 56-million-year-old plesiadapiform by Jonathan Bloch
and Doug Boyer of the University of Michigan provides important clues about primate
origins. Carpolestes simpsoni (Figure 9.8) had an opposable big toe with a flat nail,
but claws on its other digits. The claws on its feet and hands probably helped it climb
large-diameter tree trunks, but it was also able to grasp small supports. C. simpsoni
had low-crowned molars, which are suited for eating fruit. Its eyes were on the sides
of the head, and the fields of vision did not overlap. These creatures probably used
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 229
their hands and feet to grasp small branches as they climbed around in the terminal
branches of fruiting trees, and used their hands to handle fruit as they were feeding.
Experts disagree about whether plesiadapiforms ought to be included within the
primate order. They possess some but not all of the suite of traits that characterize
modern primates, and the decision about how to classify these creatures depends on
a relatively arbitrary assessment of how similar they are to other primates. The plesi-
adapiforms are important to know about, however, because they provide some infor-
mation about the traits that characterized the common ancestor of modern primates.
The discovery of C. simpsoni helps explain why natural selection favored the
basic features of primate morphology.
There are several different theories about why the traits that are diagnostic of pri-
mates evolved in early members of the primate order. Matt Cartmill, a Duke Univer-
sity anthropologist, has argued that forward-facing eyes (orbital convergence) that
provide binocular stereoscopic vision, grasping hands and feet, and nails on the toes
and fingers all evolved together to enhance visually directed predation on insects in the
terminal branches of trees. This idea is supported by the fact that many arboreal pred-
ators, including owls and ocelots, have eyes in the front of the head. However, the dis-
covery that grasping hands and feet evolved in a frugivorous plesiadapiform species
before the eyes were shifted forward presents problems for this hypothesis.
Fred Szalay of Hunter College and Marian Dagosto of Northwestern University
have suggested that grasping hands and feet and flat nails on the fingers and toes all
coevolved to facilitate a form of leaping locomotion. C. simpsoni poses a problem for
this hypothesis, too, because it had grasping hands and feet but evidently didn’t leap
from branch to branch.
Robert Sussman of Washington University hypothesized that the suite of traits that
characterize primates may have been favored because they enhanced the ability of early
primates to exploit a new array of plant resources—including fruit, nectar, flowers,
and gum—as well as insects. The early primates may have foraged and handled small
food items in the dimness of the forest night, and this nocturnal behavior may have
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 230
Sites where
FIGURE 9.9
Eocene prosimian
fossils have been found. The continents
are arranged as they were during the
early Eocene.
favored good vision, precise eye–hand coordination, and grasping hands and feet.
However, C. simpsoni foraged on fruit before orbital convergence evolved.
Finally, Tab Rasmussen, also of Washington University, has proposed that grasp-
ing hands and feet allowed early primates to forage on fruit, flowers, and nectar in the
terminal branches of angiosperms. Later, the eyes were shifted forward to facilitate
visually directed predation on insects. The idea that the evolution of grasping hands
and feet preceded the movement of the eyes to the front of the face fits with the evi-
dence from C. simpsoni.
The Eocene epoch (54 to 34 mya) was even wetter and warmer than the preced-
ing Paleocene, with great tropical forests covering much of the globe. At the beginning
of the Eocene, North America and Europe were connected, but then the two continents
separated and grew farther apart. The animals within these continents evolved in iso-
lation and became progressively more different. There was some contact between
Europe and Asia and between India and Asia during this period, but South America
was completely isolated. Primate fossils from this period have been found in North
America, Europe, Asia, and Africa (Figure 9.9). More than 200 species of fossil
prosimians have now been identified. The Eocene primates were a highly successful
and diverse group, occupying a range of ecological niches.
It is in these Eocene primates that we see at least the beginnings of all the features
that define modern primates (see Chapter 5). They had grasping hands and feet with
nails instead of claws, hind-limb-dominated posture, shorter snouts, eyes moved for-
ward in the head and encased in a bony orbit, and relatively large brains.
The Eocene primates are classified into two families: Omomyidae and Adapidae
(Figure 9.10). Although their affinities to modern primates are not known, most
researchers compare the omomyids to galagos and tarsiers and the adapids to living
lemurs (Figure 9.11). Some of the omomyids had huge eye orbits, like modern tar-
siers. This feature suggests that they were nocturnal because nocturnal primates who
do not have tapeta (singular tapetum), such as owl monkeys and tarsiers, have
extremely large orbits. (The tapetum is a reflecting layer behind the retina that
increases the light-capturing capacity of the eye.) Omomyid dentition was quite vari-
able: some seem to have been adapted for frugivory, and others for more insectivorous
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 231
diets. Some omomyids have elongated calcaneus bones in their feet, much like those
of modern dwarf lemurs, and they may have been able to leap from branch to branch.
The adapids had smaller eye orbits and were likely diurnal. They resemble living
lemurs in many aspects of their teeth, skull, nasal, and auditory regions. However, the
adapids do not display some of the unique derived traits that are characteristic of mod-
ern lemurs, including the tooth comb, a specialized formation of incisors used for
grooming. They had a range of dietary adaptations, including insectivorous, folivo-
rous, and frugivorous diets. They were generally larger than the omomyids, and their
postcranial bones (the bones that make up the skeleton below the neck) indicate that
some were active arboreal quadrupeds like modern lemurs, while others were slow
quadrupeds similar to contemporary lorises. At least one species showed substantial
sexual dimorphism, a feature that points to life in nonpair-bonded social groups.
Paleontologists are uncertain whether the modern anthropoids evolved from the
omomyids or from the adapids.
The more we learn about the primate lineages of the Eocene and early Oligocene,
the clearer it becomes that these early monkeys were extremely diverse and quite suc-
cessful. We are not entirely sure how these fossil primates are related to modern
prosimians or which of them gave rise to the first anthropoid primates.
By the end of the Eocene epoch, 34 mya, the continents were more or less positioned
on the globe as they are today. However, South America and North America were not
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 232
(a)
The behavior
FIGURE 9.11
of adapids and
omomyids probably differed. (a)
Adapids were probably diurnal. Here a
group forages for leaves. (b) Omomyids
were probably nocturnal. Several species
are shown here. (b)
yet connected by Central America; Africa and Arabia were separated from Eurasia by
the Tethys Sea, a body of water that connected the Mediterranean Sea and the Persian
Gulf. South America and Australia had completed their separation from Antarctica,
creating deep, cold currents around Antarctica (see Figure 9.4). Some climatologists
believe that these cold currents reduced the transfer of heat from the equator to
Antarctic regions, and may have been responsible for the major drop in global tem-
peratures that occurred during this period. During the Oligocene epoch, 34 to 23 mya,
temperatures dropped, and the range in temperature variation over the course of the
year increased. Throughout North America and Europe, tropical broadleaf evergreen
forests were replaced by broadleaf deciduous forests. Africa and South America
remained mainly warm and tropical.
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 233
Box 9.2
Facts That Teeth
Can Reveal
Much of what we know about the long-dead early primates and a frugivore. The insectivorous tarsier (Figure 9.14a) has
comes from their teeth. Fortunately, teeth are more durable relatively large, sharp incisors and canines, which are used
than other bones, and they are also the most common ele- to bite through the tough external skeletons of insects. In
ments in the fossil record. If paleontologists had to choose contrast, the folivorous indri (Figure 9.14b) has relatively
only one part of the skeleton to study, most would choose small incisors, and large premolars with sharp crests that
teeth. There are several reasons for this choice. First, teeth allow it to shred tough leaves. Finally, the frugivorous
are complex structures with many independent features, mangabey (Figure 9.14c) has large incisors that are used to
which makes them very useful for phylogenetic reconstruc- peel the rinds from fruit. Its molars are small because the
tion. Second, tooth enamel is not remodeled during an ani- soft, nutritious parts of fruit require less grinding than
mal’s life, and it carries an indelible record of an individual’s leaves do.
life history. Third, teeth show a precise developmental Knowing what an animal eats enables researchers to
sequence that allows paleontologists to make inferences make sensible guesses about other characteristics as well.
about the growth and development of long-dead organisms. For example, there is a good correlation between diet and
Finally, as we saw in Chapter 5, each of the major dietary body size in living primates: insectivores are generally
specializations (frugivory, folivory, insectivory) is associated smaller than frugivores, and frugivores are generally smaller
with characteristic dental features. than folivores.
Figure 9.14 diagrams one side of the upper jaw of
three modern species of primates: an insectivore, a folivore,
Sharp-pointed Large
I I
incisors incisors
I
I I
I
Large canine
C P
P C
Shearing
P P blade on
P premolars
M 3-cusped, P
P
triangular
molars P
M
M
Slicing
M
crests on
M Parallel ridges
M molars
M on molars
M
M
The right half of the upper jaws of (a) a tarsier (an insectivore), (b) an indri (a folivore),
FIGURE 9.14
and (c) a mangabey (a frugivore). I = incisor, C = canine, P = premolar, M = molar.
234
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 235
or about 5 oz). These creatures have the primitive dental formula 2.1.3.3/2.1.2.3,
which has been retained in New World monkeys but has been modified in Old World
monkeys and apes, whose dental formula is 2.1.2.3/2.1.2.3. (Dental formulas were dis-
cussed in Chapter 5.) Many aspects of parapithecid teeth and postcranial anatomy
are also primitive, suggesting that they may have been the unspecialized ancestors
both of more derived Old World monkey lineages and of the New World monkeys (see
Box 9.2).
The propliopithecids are represented by two genera and five species. These pri-
mates had the same dental formula as modern Old World monkeys, but they lack other
derived features associated with Old World monkeys. The largest and most famous of
the propliopithecids is named Aegyptopithecus zeuxis, who is known from several
skulls and a number of postcranial bones (Figure 9.15). A. zeuxis was a medium-sized
monkey, perhaps as big as a female howler monkey (6 kg, or 13.2 lb). It was a diur-
nal, arboreal quadruped with a relatively small brain. The shape and size of the teeth
suggest that it ate mainly fruit. Males were much larger than females, indicating that
they probably did not live in pair-bonded groups. Other propliopithecoids were
smaller than A. zeuxis, but their teeth suggest that they also ate fruit, as well as seeds
and perhaps gum. They were probably arboreal quadrupeds with strong, grasping feet.
Modern Old World monkeys and apes may have been derived from members of this
family.
The third group of Fayum anthropoids, the oligopithecids, were among the ear-
liest Fayum monkeys. The oligopithecids, who may have ranged beyond the Fayum
through North Africa and the Arabian Peninsula, share many primitive features with
the Eocene prosimians. However, they also share some derived features with contem-
porary anthropoids. For example, the orbits are fully enclosed in bone. The dental for-
mula of some oligopithecids was the same as that of modern Old World monkeys and
apes. It is not clear whether the reductions in the number of premolars in oligopithe-
cids and propliopithecids represent independent evolutionary events or common
ancestry.
Primates appear in South America for the first time during the Oligocene, but
it is unclear where they came from or how they got there.
The earliest New World monkey fossils come from a late Oligocene site in
Bolivia. The monkeys at this site have three premolars like modern New World mon-
keys and were about the size of owl monkeys. The shape of their molars suggests they
were frugivores. Sites in Argentina and Chile contain a number of monkey genera dat-
ing to the early and middle Miocene. They were part of a diverse animal population
Stirtonia
Lagonimico Cebupithecia
Neosaimiri
Aotus Protopithecus
Mohanamico
Homunculus
Dolichocebus
Tremacebus
Branisella
Over the last 10 years, a considerable amount of new New World fossil
FIGURE 9.16
evidence has been discovered. A number of species bear close resemblances
to modern primate species. There were once many more species of primates than we find in the same
areas today.
that included rodents, ungulates, sloths, and marsupial mammals. Most of these Pata-
gonian primates were about the size of squirrel monkeys (800 g, or 1.8 lb), though
some may have been as large as sakis (3 kg, or 6.6 lb). In Colombia, Miocene sites
dated to 12 to 10 mya contain nearly a dozen species of fossil primates. Many of these
species closely resemble modern New World monkeys (Figure 9.16). Pleistocene sites
in Brazil and the Caribbean islands have yielded a mixture of extinct and extant
species. Evidently, several species were considerably larger than any living New
World primates. Although there are no indigenous primates in the Caribbean now,
these islands once housed a diverse community of primates.
The origin of New World primates is a great mystery. It is not clear how monkeys
got to South America or how they found their way to the islands of the Caribbean. The
absence of Oligocene primate fossils in North America and the many similarities
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 237
between New World monkeys and the Fayum primates suggest to many scientists that
the ancestor of the New World monkeys came from Africa. The problem with this idea
is that we don’t know how they could have gotten from Africa to South America. By
the late Oligocene, the two continents were separated by at least 3,000 km (about
2,000 miles) of open ocean. Some authors have suggested that primates could have
rafted across the sea on islands of floating vegetation. Although there are no well-
documented examples of primates rafting such distances, fossil rodents appear in
South America about the same time and are so similar to those found in Africa that
it seems very likely that rodents managed to raft across the Atlantic.
Other researchers suggest that New World monkeys are descended from a North
American primate. But there are two problems with this hypothesis. One is that,
although there is evidence of Eurasian prosimians reaching North America earlier (dur-
ing the Eocene), there are no known anthropoid fossils from North America—for any
time period. If we assume that an early anthropoid reached North America at about
the same time, then the extensive similarities between Old World and New World mon-
keys could be readily explained. Otherwise we would have to reach the improbable
conclusion that New World monkeys are descended from a prosimian ancestor, and
that the many similarities between New World and Old World monkeys are due to
convergence.
The second difficulty with this hypothesis is geographic. Because North America
did not join South America until 5 mya, this scenario also requires the anthropoid
ancestor to have made an ocean voyage, though it may have been possible to break
up the voyage by hopping across the islands that dotted the Caribbean. Still other pos-
sibilities may exist (Figure 9.17).
The most intriguing hypothesis is that anthropoid primates actually appeared in
Africa much earlier, when a transatlantic journey would have been easier to complete.
Atlantic sea levels were lowest during the middle of the Oligocene. The Fayum pri-
mates with the closest affinities to New World monkeys are considerably younger than
this. However, this may not be a fatal liability for the hypothesis. There is good rea-
son to believe that the date of the earliest fossil we have discovered usually under-
estimates the age of a lineage. The method outlined in Box 9.3 suggests that
anthropoids actually originated at least 52 mya.
Perhaps this is how early monkeys really got from Africa to South America.
FIGURE 9.17
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 238
Box 9.3
Missing Links
University of Zürich anthropologist Robert Martin has fossil record were nearly complete, then the fact that no
pointed out that most lineages are probably older than the anthropoids living more than 35 mya have been found
oldest fossils we have discovered. The extent of the dis- would mean that they didn’t exist that long ago. However,
crepancy between the dates of the fossils that we have dis- we have reason to believe that the primate fossil record is
covered and the actual origin of the lineage depends on the quite incomplete (Figure 9.18).
fraction of fossils that have been discovered. If the primate Just as not all fossils of a given species are ever found,
Species Species
born extinct
Population size
(a)
Time
fossils formed
Number of
The sparseness
FIGURE 9.18
of the fossil
record virtually guarantees that the old-
est fossil found of a particular species
underestimates the age of the species.
(a) The population size of a hypotheti- (b)
cal primate species is plotted here Time
against time. In this case, we imagine
that the species becomes somewhat
less populous as it approaches extinc-
tion. (b) The number of fossils left by Oldest Youngest
this species is less than the population fossil fossil
fossils found
Number of
not all species are known to us either. How many species are mate order. One such tree is shown in Figure 9.19a. He then
missing from our data? Martin’s method for answering this randomly “found” 3% of the fossil species. In Figure 9.19b
question is based on the assumption that the number of the gray lines give the actual pattern of descent, and the red
species has increased steadily from 65 mya, when the first lines show the data that would be available if we knew the
primates appeared, to the present. This means that there characteristics of all the living species but had recovered
were half as many species 32.5 mya as there are now, three- only 3% of the fossil species. The best phylogeny possible,
fourths as many species 16.25 mya as there are now, and so given these data, is shown in Figure 9.19c.
on. Assuming that each species has lived about 1 million Then Martin computed the difference between the age of
years, the average life span of a mammalian species, Mar- the lineage based on this estimated phylogeny, and the actual
tin summed these figures to obtain the number of primate age of the lineage based on the original phylogeny. The dis-
species that have ever lived. Then he took the number of fos- crepancy between these values represents the error in the age
sil species discovered so far and divided that number by his of the lineage that is due to the incompleteness of the fossil
estimate of the total number of species that have ever lived. record. By repeating this procedure over and over on the com-
According to these calculations, only 3% of all fossil pri- puter, Martin was able to produce an estimate of the average
mate species have been found so far. magnitude of error, which turned out to be about 40%.
Next, Martin constructed a phylogenetic tree with the Thus, if Martin is correct, living lineages are, on average,
same number of living species that we now find in the pri- about 40% older than the age of the oldest fossil discovered.
0
Million year time period
2
4
6
8
10
12
14
16
(a) This
(a) Full tree FIGURE 9.19
tree shows
0 the complete phylogeny of a hypo-
thetical lineage. (b) This tree shows
2
Million year time period
2
species to its closest known ancestor,
4
which often means assigning inappro-
6
priate ancestors to fossil and modern
8 species, and (2) it frequently under-
10 estimates the age of the oldest mem-
12 ber of a clade.
14
16
(c) Inferred tree
239
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:17 PM Page 240
The Miocene epoch began approximately 23 mya and ended 5 mya. In the
early Miocene, the world became warmer, and once again Eurasian forests
were dominated by broadleaf evergreens like those in the tropics today. At
the end of the Miocene, the world became considerably colder and more
arid. The tropical forests of Eurasia retreated southward, and there was
more open, woodland habitat. India continued its slow slide into Asia, lead-
ing to the uplifting of the Himalayas. Some climatologists believe that the
resulting change in atmospheric circulation was responsible for the late
Miocene cooling. About 18 mya, Africa joined Eurasia, splitting the Tethys
Sea and creating the Mediterranean Sea. Because the Strait of Gibraltar had
not yet opened, the Mediterranean Sea was isolated from the rest of the
oceans. At one point, the Mediterranean Sea dried out completely, leaving
a desiccated, searing hot valley thousands of feet below sea level. About the
same time, the great north–south mountain ranges of the East African Rift
began to appear. Because clouds decrease their moisture as they rise in ele-
vation, there is an area of reduced rainfall, called a rain shadow, on the lee
Apes sometimes hang (downwind) side of mountain ranges. The newly elevated rift mountains
FIGURE 9.20
below branches while they caused the tropical forests of East Africa to be replaced by drier woodlands
feed. and savannas.
Some of the anatomical features that distinguish living apes from Old
World monkeys are related to their locomotor behavior. Monkeys climb
along on the tops of branches, using their hands and feet to grip branches,
and their tails for balance and support. They usually feed in a seated posi-
tion and have fleshy sitting pads on their bottoms. Apes make much more
use of suspensory postures during feeding and locomotion. Although they
walk on all fours when they travel on the ground, they swing from branch
to branch as they move through the trees and they often hang from
branches while feeding (Figure 9.20). Their locomotor behavior is reflected
in their skeletons. Compared to monkeys, apes have relatively short trunks,
broad chests, long arms, and flexible shoulder joints; they have no tails.
Great apes also lack fleshy sitting pads on their bottoms.
Michigan, and their colleagues resumed work in Moroto and collected additional
material (Figure 9.21). According to MacLatchy and her colleagues, M. bishopi pos-
sessed a number of skeletal features that allowed it to move like an ape, not like a mon-
key. For example, several features of the femur suggest that it might have climbed
slowly and cautiously, and the shape of the scapula indicates that it could have hung
by its arms and brachiated slowly through the trees. These features are shared with
modern apes, but not with the contemporary Miocene apes that we will meet next.
Members of
Other early Miocene apes were similar to Moropithecus in their dentition, but FIGURE 9.22
the genus Pro-
more like monkeys in their postcranial anatomy. consul were relatively large (15 to 50
kg, or 33 to 110 lb), sexually dimor-
Before Moropithecus made its debut in the pages of anthropological journals, most phic, and frugivorous. The skeleton of
paleontologists believed that the oldest hominoids were members of the family Pro- Proconsul africanus, reconstructed here,
consulidae (Figure 9.22). This family includes 10 genera and 15 species. The smallest shows that it had limb proportions much
proconsulids were about the size of capuchin monkeys (3.5 kg, or about 71⁄2 lb), and like those of modern-day quadrupedal
the largest were the size of female gorillas (50 kg, or a little over 100 lb). The pron- monkeys.
consulids seem to have occupied a range of habitats, including the tropical rain forests
in which we find apes today, and open woodlands where we now find only monkeys.
Proconsul is the best-known genus of these early hominoids. The earliest members
of this genus were found at a site called Losidok in northern Kenya and date to about 27
mya. The most recent fossils have been found at sites in Africa dated to 17 mya. Plant
and animal fossils found with Proconsul indicate that they lived in rain-forest environ-
ments like those found in tropical Africa today. Proconsul species share several derived
features with living apes and humans that we don’t see in anthropoid primates. For exam-
ple, Proconsul didn’t have a tail and did not have the fleshy sitting pads that Old World
monkeys and gibbons have. Proconsul species also had somewhat larger brains in rela-
tionship to body size than similarly sized monkeys had. Otherwise, members of the genus
Proconsul were similar to Aegyptopithecus and the other Oligocene primates. Their teeth
had thin enamel, which is consistent with a frugivorous diet. Their postcranial anatomy,
including the relative length of their arm and leg bones and narrow and deep shape of
their thorax, were much like that of quadrupedal monkeys, but certain features of their
feet and lower legs were more apelike. Proconsul had an opposable thumb, a feature that
we see in humans but not in living apes or monkeys. All of the Proconsul species show
considerable sexual dimorphism, suggesting that they were not pair-bonded.
Besides Moropithecus and Proconsul, there were a number of other early Miocene
apes in Africa. These creatures have derived features of apes in their faces and teeth,
but not in their postcranial anatomy (Figure 9.23).
The middle Miocene epoch saw a new radiation of hominoids and the expansion
of hominoids throughout much of Eurasia.
Exploration of middle Miocene (15 to 10 mya) deposits has yielded a great abun-
dance of new hominoid species in Africa, Europe, and Asia. Examples include
Kenyapithecus and Chororapithecus from what is now East Africa, Oreopithecus (Fig-
ure 9.24), Dryopithecus, and Ouranopithecus from Europe, and Lufengpithecus and
Sivapithecus from Asia. The skulls and teeth of these hominoids typically differ from
those of the proconsulids in a number of ways that indicate they ate harder or more
fibrous foods than their predecessors. Their molars had thick enamel for longer wear,
and rounded cusps, which are better suited to grinding. Their zygomatic arches
(cheekbones) flared farther outward to make room for larger jaw muscles. And the
lower jaw was more robust, to carry the forces produced by those muscles. It seems
likely that these features were a response to the climatic shift from a moist, tropical
environment to a drier, more seasonal environment with tougher vegetation and
harder seeds.
Until recently, little was known about the postcranial skeletons of middle Miocene
apes. However, two spectacular discoveries by a team led by Salvador Moyà-Solà and
Meike Köhler of the Institut de Paleontologia Miquel Crusafont, Spain, from sites near
Barcelona have done much to change this. In 2004, the team reported the discovery
of a remarkably complete skeleton a new species, Pierolapithecus catalaunicus, which
shows some, but not all, of the derived morphological traits of modern great apes (Fig-
10 cm ure 9.25). The fossil includes well-preserved cranial, dental, and postcranial material
from a single individual that lived about 13 mya. This creature had a small, flattened,
Oreopithecus apelike face as well as a number of morphological features associated with upright pos-
FIGURE 9.24
had a number ture and locomotion. The wrist is flexible, the rib cage is wide and shallow, the scapula
of traits that are associated with sus- is shifted to the back, and the lumbar region of the spine is short and stiff. However,
pensory locomotion, including a rela-
tively short trunk, long arms, short legs,
long and slender fingers, and great
mobility in all joints. The phylogenetic
affinities of this late-Miocene ape from
Italy are not well established.
the finger bones are not as curved as the fingers of orangutans. Moyà-Solà and his col-
leagues interpret this to mean that they were not adapted for suspensory locomotion.
If this interpretation is correct, then the traits associated with suspensory locomotion
must have evolved independently in gibbons and the great apes, after they diverged
from a common ancestor about 14.5 mya.
Strong evidence for suspensory locomotion in Miocene apes comes from another
site near Barcelona excavated by Moyà-Solà, Köhler, and their colleagues. There they
discovered a remakably complete specimen of Dryopithecus dated 9.5 mya (Figure
9.26). The anatomy of the bones clearly indicates that these creatures moved through
the canopy by swinging from one branch to another and were adapted for suspensory
postures.
There are no clear candidates for the ancestors of humans or any modern apes
except perhaps orangutans.
The evolutionary history of the apes of the Miocene is poorly understood. There
were many different species, and the phylogenetic relationships among them remain
largely a mystery. The recent revelations about Morotopithecus and the discovery of
Pierolapithecus only add to the uncertainty. We have no clear candidates for the ances- Much of the
tors of any modern apes except for the orangutan, who shares a number of derived
FIGURE 9.26
skeleton of
skull features with Sivapithecus of the middle Miocene. The teeth of the newly dis- Dryopithecus was found at a site in
covered Chororapithecus bear similarities to modern gorillas, but it is not clear Spain dated to 9.5 mya. The character-
whether these similarities reflect common ancestry or convergence. istics of the skeleton indicate that it was
We can be almost certain that the earliest hominins evolved from some type of adapted for suspensory postures, much
Miocene ape, but we have no idea which one it was. Nonetheless, the study of ape and like modern apes are.
hominin origins in the Miocene has been very useful for paleontologists. John Fleagle
of Stony Brook, State University of New York, points out that, in the process of con-
sidering and ultimately rejecting many different candidates for the “missing link”
between apes and early hominins, paleoanthropologists have come to realize that the
key features distinguishing the earliest hominins from ancestral hominoids were not
the features that they had originally been looking for: the big brains and small teeth
that characterize modern humans. Instead, it was a suite of skeletal adaptations for
bipedalism that marked the primary hominin adaptation.
During the early and middle Miocene, ape species were plentiful and monkey
species were not. In the late Miocene and early Pliocene, many ape species
became extinct and were replaced by monkeys.
Apes flourished during the Miocene, but all but a few genera and species eventu-
ally became extinct. Today there are only gibbons, orangutans, gorillas, and chim-
panzees. We don’t know why so many ape species disappeared, but it seems likely that
many of them were poorly suited to the drier conditions of the late Miocene and early
Pliocene.
The fossil record of Old World monkeys is quite different from that of the apes.
Monkeys were relatively rare and not particularly variable in the early and middle
Miocene, but the number and variety of fossil monkeys increased in the late Miocene
and early Pliocene. Thus, although there are many more extinct ape species than liv-
ing ones, the number of living monkey species greatly exceeds the number of extinct
monkey species.
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:18 PM Page 244
Once again, the fossil record reminds us that evolution does not proceed on a
steady and relentless path toward a particular goal. Evolution and progress are not syn-
onymous. During the Miocene, there were dozens of ape species but relatively few
species of Old World monkeys. Today there are many monkey species and only a hand-
ful of ape species. Despite our tendency to think of ourselves as the pinnacle of evo-
lution, the evidence suggests that, taken as a whole, our lineage was poorly suited to
the changing conditions of the Pliocene and Pleistocene.
FURTHER READING
1. Briefly describe the motions of the continents over the last 180 million years. Why
are these movements important to the study of human evolution?
2. What has happened to the world’s climate since the end of the Age of Dinosaurs?
Explain the relationship between climate change and the notion that evolution
leads to steady progress.
3. What are angiosperms? What do they have to do with the evolution of the
primates?
4. Why are teeth so important for reconstructing the evolution of past animals?
Explain how to use teeth to distinguish among insectivores, folivores, and
frugivores.
5. Which primate groups first appear during the Eocene? Give two explanations for
the selective forces that shaped the morphologies of these groups.
6. Why is there a problem in explaining how primates arrived in the New World?
7. Why does the oldest fossil in a particular lineage underestimate the true age of the
lineage? Explain how this problem is affected by the quality and completeness of
the fossil record.
8. Explain how potassium–argon dating works. Why can it be used to date only vol-
canic rocks older than about 500,000 years?
09_Tree Shrew to Ape [4p]:How Humans Evolved [5e] 10/10/08 1:18 PM Page 245
KEY TERMS
Chapter 10
At the Beginning
Sahelanthropus tchadensis
Orrorin tugenensis
Ardipithecus
The Hominin Community Diversifies
Australopithecus
A. anamensis
A. afarensis
A. africanus
A. garhi
Paranthropus
Kenyanthropus
Homo habilis/rudolfensis
Hominin Phylogenies
The Evolution of Early Hominin
Morphology and Behavior
The Evolution of Bipedalism
Early Hominin Subsistence
Early Hominin Social Organization
D
uring the Miocene, the Earth’s temperature began to fall. This global cool-
ing caused two important changes in the climate of the African tropics.
First, the total amount of rain that fell each year declined. Second, rainfall
became more seasonal, so there were several months each year when no rain fell. As
the tropical regions of Africa became drier, moist tropical forests shrank and drier
woodlands and grasslands expanded. Like other animals, primates responded to these
ecological changes. Some species, including many of the Miocene ape lineages,
apparently failed to adapt and became extinct. The ancestors of chimpanzees and
gorillas remained in the shrinking forests and carried on their lives much as before.
246
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 247
Changes brought about through generations of natural selection allowed a few species
to move down from the trees, out of the rain forests, and into the woodlands and
savannas. Our ancestors, the earliest hominins, were among these pioneering species.
The spread of woodland and savanna led to the evolution of the first hominins
about 6 mya.
Beginning about 6 mya, hominins begin to appear in the fossil record. Between 4
and 2 mya, a diverse community of hominin species ranged through eastern and south-
ern Africa. These creatures were different from any of the Miocene apes in two ways:
First, and most important, they walked upright. This shift to bipedal locomotion led
to major morphological changes in their bodies. Second, in the new savanna and
woodland habitats, new kinds of food became available. As a result, the hominin
chewing apparatus—including many features of the teeth, jaws, and skull—changed.
Otherwise, the behavior and life history of the earliest hominins were probably not
much different from those of modern apes.
Because these early hominins were bipedal and shared many dental features with mod-
ern humans, they are classified as members of our subfamily. However, because their
brains and life history were very similar to those of contemporary chimpanzees, early
hominins are not included in the same genus (Homo) as modern humans.
In this chapter, we describe the species that constituted this early hominin com-
munity, and we discuss the selective forces that transformed an ancestral, arboreal ape
that was something like an orangutan or chimpanzee into a diverse community of
bipedal apes living in the woodlands and savannas of Pliocene Africa. In subsequent
chapters, we will consider how one of these savanna apes became human.
Many of the morphological features that we focus on in this chapter may seem
obscure, and you may wonder why we spend so much time describing them. These fea-
tures help us identify hominin species and trace the origins of traits that we see in later
species. We can also use some of these characteristics to reconstruct aspects of diet,
social organization, and behavior.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 248
AT THE BEGINNING
Genetic data indicate that the last common ancestor of humans and chimpanzees lived
between 7 and 5 mya. Until recently, we knew very little about the hominin lineage
during this period. In the last few years, however, fossil discoveries of several kinds of
creatures—Sahelanthropus tchadensis, Orrorin tugenensis, Ardipithecus kadabba, and
Ardipithecus ramidus—have begun to shed light on this important period in the his-
tory of the human lineage. During this period, we see the first hints of some of the dis-
tinctive features that differentiate hominins from apes—some evidence of bipedal
locomotion (see Box 10.1), large posterior teeth, and canine reduction. However, the
fossil materials are fragmentary, the combinations of these traits vary, and the phylo-
genetic relationships between these creatures and other hominins are unclear.
Sahelanthropus tchadensis
In the search for the oldest members of the human lineage, most researchers have con-
centrated their efforts on the Rift Valley of East Africa, where many important fossil
discoveries have been made over the last 50 years. So, in 2002, the world of paleon-
tology was stunned when a team of researchers led by Michel Brunet of the Univer-
sity of Poitiers in France announced a remarkable find from a site in Chad, 1,500 miles
(2,500 km) east of the Rift Valley. The fossil material consists of a nearly complete cra-
nium (the skull minus the lower jawbone), four partial mandibles, and four teeth (Fig-
ure 10.4). Brunet and his colleagues gave the fossil the scientific name Sahelanthropus
tchadensis. The word Sahelanthropus comes from “Sahel,” the vast dry region south
of the Sahara; and tchadensis comes from “Chad,” the country where the fossil was
found. Although the site at which Sahelanthropus tchadensis was found is now a bar-
ren desert, it was much different millions of years ago. Lakes and rivers sustained
forests and wooded grasslands.
This fossil shook up the paleontological community for a number of reasons.
First, since Chad is far from other paleontological sites in East Africa, the discovery
implies that hominins had a much larger range than was previously believed. Second,
this fossil is surprisingly old. The geology of the site does not allow radiometric or
paleomagnetic dating. However, there is a close match between the other fossil ani-
mals found at the site and the fauna found at two sites in East Africa securely dated
to between 7 and 6 mya. This makes Sahelanthropus tchadensis the oldest hominin
cranium. Sahelanthropus tchadensis may give us a picture of the last common ances-
tor because genetic data tell us that the hominin lineage diverged from the chim-
panzee lineage during this period of time. Third, and most important, is the fact that
it possesses a very surprising mix of anatomical features. The foramen magnum, the
hole in the skull through which the spinal cord passes, is located under the skull,
rather than at the back—a feature that is associated with bipedal locomotion.
Because of this, Brunet and his colleagues think that these creatures walked upright.
However, their brains were no bigger than the brains of chimpanzees. Sahelanthro-
pus tchadensis’s brains were 320 to 350 cc (cubic centimeters); for comparison,
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 249
BOX 10.1
What It Takes to
Be a Biped
Bipedal locomotion distinguishes hominins from hominoids. But your torso does not tip, because the torque is opposed
The transition from a forest ape to a terrestrial biped by abductors, muscles that run from the outer side of the
involves many new adaptations. A number of these changes pelvis to the femur. At the appropriate moment during each
are reflected in the morphology of the skeleton. Thus, by stride, these muscles tighten and keep you upright. (You can
studying the shape of fossils, we can make inferences about demonstrate this by walking around with your open hand
the animal’s mode of locomotion. Changes in the pelvis pro- on the side of your hip. You’ll feel your abductors tighten as
vide a good example. In terms of shape and orientation, the you walk and your torso tip if you relax these muscles. You
human pelvis is very different from that of forest-dwelling might want to do this in private.) The abductors are
apes like the chimpanzee (Figure 10.1). attached to the ilium (plural ilia), a flaring blade of bone on
Other changes are more subtle but also diagnostic. the upper end of the pelvis). The widening and thickening of
When modern humans walk, a relatively large proportion of the ilium and the lengthening of the neck of the femur (the
the time is spent balanced on one foot. Each time you take thighbone) add to the leverage that the abductors can exert
a step, your body swings over the foot that is on the ground, and make bipedal walking more efficient. In addition, the
and all of your weight is balanced over that foot. At that distribution of cortical bone in the femur is diagnostic of
moment, the weight of your body pulls down on the center locomotor patterns. In humans the cortical bone is thickest
of the pelvis, well inward from the hip joint (Figure 10.2). along the lower edge of the femoral neck, while in chim-
This weight creates a twisting force, or torque, that acts to panzees cortical bone is evenly distributed on the upper and
rotate your torso down and away from the weighted leg. lower edge.
Front Top
Front Top
Australopithecus afarensis
Chimpanzee
Homo sapiens
249
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 250
A schematic
FIGURE 10.2 Pelvis Pelvis
diagram of the
lower body at the point of the stride in
which all of the weight is on one leg.
Body weight acts to pull down through Body Unweighted Body Weighted
the centerline of the pelvis. This creates Unweighted weight Weighted leg weight leg
a torque, or twisting force, around the leg leg
hip joint of the weighted leg. (a) If this
torque were unopposed, the torso would
twist down and to the left. (b) During
each stride the abductor muscles tighten
to create a second torque that keeps
the body erect. (a) Without abductors (b) With abductors
The modern human knee joint is quite different from and the end of the femur at the knee joint is not slanted. The
the chimpanzee knee joint (Figure 10.3). Efficient bipedal feet also show a number of derived features associated with
locomotion requires the knees to lie close to the centerline bipedal locomotion, including a longitudinal arch and a
of the body. As a result, the human femur slants down and humanlike ankle.
inward, and its lower end is angled at the knee joint to make The presence or absence of these features in fossil skele-
proper contact with the bones of the lower leg. In contrast, tal material allows paleontologists to make strong inferences
the chimpanzee femur descends vertically from the pelvis, about the model of locomotion that the animals used.
250
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 251
chimpanzee brains are about 400 cc. The teeth are different
than the teeth of chimpanzees in several ways: the canines
are smaller, the upper canine is not sharpened against the
lower premolar as it is in chimpanzees, and the enamel is
thicker. The face is relatively flat, and there is a massive
brow ridge over the eyes. These are features associated with
hominins who date to later than 2 mya but not with chim-
panzees or with the australopiths, the hominins who dom-
inate the fossil record between 4 and 2 mya. This has
caused some authorities to doubt that Sahelanthropus is a
human ancestor.
Orrorin tugenensis
In the spring of 2001, a French–Kenyan team led by Brigitte The cranium of Sahelanthropus tchadensis
Senut of the National Museum of Natural History in Paris FIGURE 10.4
found in Chad dates to between 7 and 6
and Martin Pickford of the Kenya Palaeontology Expedition mya. It has a flat face and large browridge. Australopiths dated to
announced the discovery of 12 hominin fossils in the Tugen between 4 and 2 mya have more apelike prognathic faces and lack
Hills in the highlands of Kenya. These fossils, which include browridges. (Photograph courtesy of Michel Brunet.)
parts of thigh and arm bones, a finger bone, two partial
mandibles (lower jaws), and several teeth, are securely dated to 6 mya (Figure 10.5).
Senut and Pickford assigned their finds to a new species, Orrorin tugenensis. The genus
name means “original man” in the language of the local people, and it reminded Senut
and Pickford of the French word for “dawn,” aurore. Fossils of forest creatures, such
as colobus monkeys, and open-country dwellers, such as impala, were found in the
same strata as O. tugenensis, indicating that the habitat was a mix of woodland and
savanna.
Like the fossils of Sahelanthropus, these specimens are similar to chimpanzees in
some ways and to humans in others. The incisors, canines, and one of the premolars
are more like the teeth of chimpanzees than of later hominins. The molars are
smaller than those of A. ramidus and later apelike hominins, and they have thick
enamel like human molars have. As in chimpanzees and some later hominins, the arm
and finger bones have features that are believed to be adaptations for climbing. Senut
and her colleagues argue that several features of the femur (thighbone), such as the
long neck that connects the shaft to the head of the femur, are more similar to those
of later bipedal hominins than to those of quadrupedal apes. The research team has
also used computerized tomography (CT) to assess the distribution of cortical bone
in the Orrorin tugenensis femur. In modern humans, the cortical bone is thicker along
the lower edge of the femoral neck than it is along the upper edge, while in chim-
panzees the cortical thickness is the same in both. Senut and her colleagues argue that
the distribution of cortical bone in Orrorin tugenensis is more humanlike than ape- The fossils of
like, which suggests that these creatures were bipedal. Not all experts are convinced
FIGURE 10.5
Orrorin tuge-
by these analyses, and the issue will not be settled until additional skeletal material is nensis include parts of the femur, lower
recovered. jaw, a finger bone, and teeth.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 252
Ardipithecus
Some of the most important evidence of human origins comes from the northern end
of the Rift Valley, in the middle Awash region of Ethiopia. Although this area is now
dry and desolate, it was once the site of woods and grasslands. Here, many crucial
events in the history of the human lineage unfolded, and a number of spectacular dis-
coveries of hominin ancestors have been made.
About 5.8 to 5.2 mya, this region was occupied by an apelike hominin named
Ardipithecus ramidus Ardipithecus kadabba. This species was discovered by Yohannes Haile-Selassie,
who was then a graduate student at the University of California, Berkeley. The finds,
which consist mainly of teeth and a single foot bone, were made at a site called
Aramis, in the Middle Awash basin. Like Sahelanthropus, Ardipithecus kadabba pos-
sesses a mixture of primitive and derived dental traits. For example, the canine sharp-
ens itself against the first premolar, a trait that is found in chimpanzees but not in
modern humans. On the other hand, Ardipithecus kadabba has thicker enamel on its
molars than chimpanzees do, and the shape of its canines is like that of hominins that
Ardipithecus kadabba succeeded Ardipithecus kadabba in the area (Figure 10.6). A single toe bone dated
to 5.2 mya is similar to the toe bones of other bipedal hominins and may be diag-
nostic of bipedal locomotion. Again, more postcranial material is needed to confirm
this interpretation.
About a million years after Ardipithecus kadabba was wandering around the
Awash, another member of the same genus appears in the fossil record. These crea-
tures were discovered in 1992 by the members of an expedition led by Tim White
of the University of California, Berkeley. These fossils, dated to 4.4 mya, include
teeth and jaws, the lower part of the skull, and parts of the upper arms. White and
his colleagues, Gen Suwa of the University of Tokyo and Berhane Asfaw of Addis
Pan troglodytes A. kadabba Ababa, named the species Ardipithecus ramidus, from the words ardi (meaning
“ground” or “floor”) and ramis (meaning “root”) in the local Afar language. In
In Ardipithecus
FIGURE 10.6 2005, a team led by Sileshi Semaw of Indiana University announced the discovery
kadabba and
of fossils representing at least 9 more members of the same species from Gona, an
chimpanzees, the canines are sharpened
area not far from Aramis.
against the first premolar. This is not the
case with later australopiths (top and Several features suggest that Ardipithecus ramidus was a hominin. The foramen
middle art). However, the shape of the magnum is located forward under the skull, as it is in humans, rather than at the back
canine of Ardipithecus kadabba is similar of the skull, as it is in chimpanzees. Ardipithecus ramidus also shares several other
to Ardipithecus ramidus and Ardipithe- derived features with later hominins. The upper canine is not sharpened by the lower
cus afarensis, which follow Ardipithecus premolar, as it is in chimpanzees, and the canine is shaped more like incisor teeth than
kadabba. (Redrawn from Y. Haile- in chimpanzees. Ardipithecus ramidus also resembles humans in several additional fea-
Selassie, G. Suwa, and T. D. White, tures of the teeth and arm bones. Other traits are much more apelike. The molars are
2004, “Late Miocene Teeth from smaller in relation to body size than in other early hominins, the enamel is thinner than
Middle Awash, Ethiopia, and Early in other early hominins, and the canines are smaller than in chimpanzees and gorillas
Hominid Dental Evolution,” Science
but larger than in later hominins. In addition, the deciduous (baby) molars and jaw
303:1503–1505.)
joint are very similar to those of apes, and the base of the skull is more pneumatized
(filled with air pockets) than in later hominins.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 253
Ardipithecus
FIGURE 10.7
ramidus fossils
THE HOMININ COMMUNITY DIVERSIFIES are found in association with the fossils
of woodland species like this kudu, sug-
gesting that Ardiphithecus ramidus lived
A number of hominin species lived in Africa between 4 and 2 mya. They in a forested environment.
are divided into four genera: Australopithecus, Paranthropus, Kenyanthropus,
and Homo.
Beginning about 4 mya the hominin lineage proliferated, and over the next 2 million
years there were between four and seven species of hominins living in Africa at any
given time. A number of different taxonomic schemes have been used to classify these
creatures, although there is no consensus about their phylogenetic relationships. We
follow the following taxonomic scheme:
1. Australopithecus includes four species: A. anamensis, A. afarensis, A. africanus,
and A. garhi. The genus name means “southern ape” and was first applied to an
A. africanus skull found in South Africa. The species were small bipeds with
teeth, skull, and jaws adapted to a generalized diet.
2. Paranthropus includes three species: P. aethiopicus, P. robustus, and P. boisei. The
genus name means “parallel to man” and was coined by Robert Broom, who dis-
covered the first specimens of P. robustus. These species were similar to the mem-
bers of the genus Australopithecus from the neck down, but they had massive teeth
and jaws adapted to heavy chewing of tough plant materials, and a skull modi-
fied to carry the enormous muscles necessary to power this chewing apparatus.
3. Kenyanthropus includes only one species: K. platyops. The genus name means
“Kenyan man” and was given to a specimen found in northern Kenya. We do
not yet know much about these creatures, but they are distinguished by a flat-
tened face and small teeth.
4. The first members of the genus Homo probably coexisted with several other
hominin species in East Africa. They had larger brains and smaller teeth than
contemporary hominin species.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 254
H. ergaster
P. boisei
Omo
P. aethiopicus
A. afarensis
H. ergaster
H. habilis
H. rudolfensis
Lake Turkana P. boisei
K. platyops
P. aethiopicus
A. anamensis
Tugen Hills Orrorin tugenensis
H. ergaster
Olduvai P. boisei
H. habilis
Laetoli A. afarensis
Uraha H. rudolfensis
H. ergaster
Hominin fossils
FIGURE 10.8 P. robustus
have been dis- South Africa
covered at a number of sites in eastern H. habilis
and southern Africa. Fossils belonging A. africanus
to each genus have been assigned a dif-
ferent color. Fossils discovered at Bahr
el Ghazal in Chad have been provision-
ally assigned to the species A. afarensis, The sites at which early hominin fossils have been found in Africa are identified on the
B map in Figure 10.8. In this section, we briefly describe the history and characteristics
pending more detailed analysis.
of each of these fossil species and then turn to a discussion of their common features.
You are excused if you feel as though you have mistakenly picked up a Russian
novel with a long cast of characters filled with tongue twisting, hard-to-remember
names. Keep in mind, however, that all of these species share some important char-
acteristics. They were bipedal on the ground but were probably also good tree
climbers; their brains were about the same size as the brains of modern apes; they had
smaller canines and incisors but bigger molars and premolars with thicker enamel than
chimpanzees. They were considerably smaller than modern humans, and sexual
dimorphism was pronounced. These similarities have led many anthropologists to refer
to all of these creatures, except Homo, collectively as “australopiths,” a usage we will
adopt here.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 255
Australopithecus
A. ANAMENSIS
Australopithecus anamensis was bipedal but had a more apelike skull than later
australopithecines had.
A. AFARENSIS
of a lower jaw from Bahr el-Ghazal in Chad. This fossil, provisionally identified as
A. afarensis, is between 3.1 and 3.4 million years old, judging by the age of associated
fossils. In South Africa, Ronald Clarke of the Johann Wolfgang Goethe University of
Frankfurt, Germany, and Phillip Tobias of the University of the Witwatersrand,
Johannesburg, found several foot bones that may also belong to A. afarensis.
Reconstructions of the environments at these sites indicate that A. afarensis lived
in habitats ranging from woodland to dry savanna. Between 4 and 3 mya, the envi-
ronment at Hadar was a mix of woodland, scrub, and grassland; at Laetoli, it was a
dry grassland sparsely dotted with trees.
The cranium of A. afarensis is quite apelike, with a small brain and powerful
chewing capacities. Its endocranial volume (the capacity of its brain cavity) is about
450 cc—less than a pint. This is about the same size as the brain of a modern chim-
panzee. The cranium of A. afarensis also shows many other apelike cranial charac-
teristics. For example, the base of the cranium is flared at the bottom, and the bone is
pneumatized; the front of the face below the nose is pushed out—a condition known
as subnasal prognathism; and the jaw joint is shallow (Figure 10.15). There are also
Here, Lucy’s
many apelike features of the dentition, including substantial sexual dimorphism in FIGURE 10.13
skeleton
canine size, relatively large and procumbent (forward-slanting) incisors, and the pres-
(left) stands beside the skeleton of a
ence of a diastema (space between the upper canine and incisor that accommodates modern human female. The parts of the
the lower canine) in many specimens (Figure 10.16). skeleton that have been discovered are
The Dikika child’s skeleton included an intact hyoid bone. The hyoid is a small shaded. Lucy was shorter than modern
bone that connects the muscles of the tongue to the muscles of the larynx. The females and had relatively long arms and
anatomy of the throat affects the range of sounds that can be made. The shape of the a relatively small brain. (Art: © 1985
David L. Brill.)
Temporal
line
Nuchal Chimpanzee,
line Pan troglodytes
Subnasal
prognathism
Broad (heavily pneumatized) cranial base
Diastema
The cranium
FIGURE 10.15
of Australo- Australopithecus
pithecus afarensis possesses a number afarensis
of primitive features, including a small
brain, a shallow jaw joint, a pneumatized
cranial base, and subnasal prognathism
in the face. (Figure courtesy of Richard
Klein.)
Diastema Diastema
(No diastema)
A trail of fossil footprints proves that a striding biped lived in East Africa at
the same time as Australopithecus afarensis.
The conclusion that A. afarensis had an efficient striding gait was dramatically
strengthened by a remarkable discovery by Mary Leakey and her coworkers at Lae-
toli in Tanzania. They uncovered a trail of footprints 30 m (about 100 ft) long that
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 260
had been made by three bipedal individuals who had crossed a thick bed of wet vol-
canic ash about 3.5 mya (Figure 10.18). Paleontologists estimate that the tallest of the
individuals who made these prints was 1.45 m (4 ft 9 in.) and the shortest was 1.24
m (4 ft 1 in.) tall. The path of these three individuals was preserved because the wet
ash solidified as it dried, leaving us a tantalizing glimpse of the past. University of
Chicago anthropologist Russell Tuttle has compared the Laetoli footprints with the
footprints made by members of a group of people called the Machigenga, who live in
the tropical forests of Peru. Tuttle chose the Machigenga because their average height
is close to that of the makers of the Laetoli footprints, and they do not wear shoes. Tut-
tle found that the Laetoli footprints are functionally indistinguishable from those made
by the Machigenga, and he concluded that the creatures who made the footprints
walked with a fully modern striding gait.
Who made these footprints? The prime suspects are Australopithecus afarensis
Several and Kenyanthropus platyops (a hominin you will meet later in this chapter) because
FIGURE 10.18
bipedal crea- they are the only hominins known to have lived in East Africa at the time the tracks
tures walked across a thick bed of wet were made. Australopithecus afarensis is the most likely culprit because it is the only
volcanic ash about 3.5 mya. Their foot- hominin whose remains have been found at Laetoli. If A. afarensis did make the foot-
prints were preserved when the ash prints, then the researchers who doubt that they were striding bipeds are wrong. How-
solidified as it dried. ever, some anthropologists are convinced by the anatomical evidence that A. afarensis
was not a modern biped. If they are correct, then K. platyops made the footprints, or
there was another, as yet undiscovered, hominin species living in East Africa 3.5 mya.
All nonhuman primates, except some gorillas, spend the night perched in trees or
huddled on cliffs to protect themselves from nocturnal predators like leopards. Even
chimpanzees, who are about the size of Lucy, make leafy nests and sleep in trees each
night. A number of features of the skeleton of A. afarensis suggests that they may have
slept and foraged in trees, too. In chimpanzees, the humerus, the radius and ulna (the
lower bones of the arm), and the femur are about the same length; in modern
humans, the bones of the forearm have become shorter and the femur has become
longer than the humerus. Australopithecus afarensis resembles chimpanzees in the pro-
portions of these bones. In addtion, the bones of their fingers and toes curved like those
of modern apes. Their thumbs were short and the tips of their fingers tapered, while
modern humans have longer thumbs and broader fingertips. These traits would have
made A. afarensis well suited to grasping branches as it climbed. Certain distinctive
features of their scapulae (shoulder blades) are also well suited to supporting the body
in a hanging position.
The well-preserved scapulae of the Dikika child have added to the controversy
about the arboreal adaptations of A. afarensis. The shoulder socket faces upward as
in apes, not to the side as in modern humans. This may facilitate the kinds of move-
ments that apes use when they are climbing. Alemseged and his colleagues think that
the Dikika child’s scapula is more similar to the scapula of modern gorillas, which
spend most of their time on the ground, than to the scapula of chimpanzees, which are
partly arboreal (Figure 10.19).
Anatomical evidence suggests that there was considerable variation in the body
size of A. afarensis adults in the Hadar population. The bigger individuals were 1.51
m (5 ft) tall and weighed about 45 kg (100 lb). The smaller ones were about 1.05 m
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:07 PM Page 261
(31⁄2 ft) tall and weighed about 30 kg (65 lb). Thus, the big ones
were about 1.5 times larger than the small ones. There are two
possible explanations for variation in the Hadar population. The
variation could represent sexual dimorphism, with the larger
adults being male and the smaller ones female. The difference
between the Hadar males and females approaches the magnitude
of sexual dimorphism in modern orangutans and gorillas and is
considerably greater than that in modern humans, bonobos, and
chimpanzees.
On the other hand, it is possible that large and small individ-
uals represent two different species. This interpretation was ini-
tially supported by the fact that larger individuals were much more
common at Laetoli, which was then a savanna, while the smaller
ones were found mainly in what were then forested environments.
Some researchers also believe that the smaller individuals were less
suited to bipedal locomotion than the bigger ones were. This sug-
gests that one species was big, fully bipedal, and adapted for
savanna life. The morphology of the smaller species represented a The shape and orientation of the
compromise between tree climbing and walking—an adaptation FIGURE 10.19
scapula (shoulder) of the Dikika
for life in the forest. child (a) is compared to the scapula of a gorilla (b), modern
The fossils discovered in Ethiopia by Tim White and Donald human (c), and chimpanzee (d) of approximately the same age.
Johanson suggest that the two-species theory is wrong. The new The green line indicates where the spine meets the scapula. It
fossils encompass the full range of variation in size at a single site is somewhat surprising to find that the scapula of the Dikika
and over a fairly narrow range of dates. Moreover, a large femur child looks more like the scapula of gorillas than chimpanzees,
shows many of the same features seen in Lucy’s more petite femur. because gorillas use suspensory postures less than chimpanzees
The new data also suggest that both large and small forms were do. (From Z. Alemseged, F. Spoor, W. H. Kimbel, R. Bobe,
D. Geraads, D. Reed, and J. G. Wynn, 2006, “A Juvenile
found in the full range of environments from forest to savanna. All
Early Hominin Skeleton from Dikika, Ethiopia,” Nature
of these facts suggest that this assemblage of fossils represents a
443:296–301.)
single, sexually dimorphic species. Most paleoanthropologists
now accept this conclusion, although there is some dispute about whether the degree
of dimorphism was large like modern gorillas or orangutans, or smaller, more like
chimpanzees and humans.
A. AFRICANUS
Australopithecus africanus is known from several sites in South Africa that date
from 3 to 2.2 mya.
Deciduous
The first australopithecine specimen was identified canines
FIGURE 10.20
in South Africa by Raymond Dart in 1924. Dart
named it Australopithecus africanus, which means “the southern ape of 0 5 cm
Africa,” but the specimen is often called the Taung child. Dart concluded that
the Taung child was bipedal. Even though it had a very small brain, he believed
it to be intermediate between humans and apes. His conclusions were not Permanent first
generally accepted for nearly 30 years. (Figure courtesy of Richard Klein.) molars (erupting)
those days most physical anthropologists believed that large brains had evolved in the
human lineage before bipedal locomotion. Controversy about the taxonomic status of
A. africanus continued for the next 30 years.
The creature who caused all the controversy was a small biped with relatively
modern dentition and postcranial skeleton. As with A. afarensis, there is pronounced
sexual dimorphism, in both canine and body size. Males stood 1.38 m (4 ft 6 in.) tall
and weighed 41 kg (90 lb), and females were 1.15 m (3 ft 9 in.) tall and tipped the
scales at about 30 kg (66 lb). The teeth are more modern than those of A. afarensis
in several ways. The brain averaged about 460 cc, somewhat larger than the average
for A. afarensis. However, the difference in cranial capacity between A. afarensis and
A. africanus is small compared with that among individuals and is likely due to sam-
pling variation. Members of this species lived from 3.0 to 2.2 mya. The postcranial
skeleton is virtually identical to that of A. afarensis. Dart’s notion that the Taung child
was bipedal was controversial because it relied primarily on the location of the fora-
men magnum, and he had no postcranial bones to corroborate his conclusions. Sub-
sequently, many adult skulls and postcranial bones were found at two other sites in
South Africa: Makapansgat and Sterkfontein. The hip bone, pelvis, ribs, and vertebrae
of A. africanus are much like those of Lucy and the First Family. Dart’s claim that the
Taung child was bipedal has been strongly supported.
New methods for analyzing dental enamel provide clues about what
Astralopithecus africanus ate.
While the old saw “You are what you eat” might not be right in general, it is true
of your tooth enamel. To understand this, we need to do a little bit of botany and
chemistry. Plants use one of two chemically distinct kinds of photosynthesis. Woody
plants like trees, bushes, and shrubs utilize one type, which is called C3 photosynthe-
sis, while grasses and sedges use a second type, called C4 photosynthesis. C4 plants
have higher concentrations of the heavy isotope of carbon, 13C, than do C3 plants. Ani-
mals that eat plants (or eat things that eat plants) incorporate the carbon isotopes into
the enamel of their teeth, and the ratio of the two carbon isotopes, 13C versus 12C, in
tooth enamel tells us something about the kinds of foods the animals ate. Matt Spon-
heimer, now at the University of Colorado at Boulder, and Julia Lee-Thorpe, now at
the University of Bradford, have used stable isotope mass spectrometry to analyze the
composition of tooth enamel in A. africanus and a number of other mammalian taxa
from several sites in South Africa. The carbon isotope values for A. africanus indicate
that its diet was more variable than the diets of all but one of the other mammalian
species sampled and included substantial amounts of C4 foods. This suggests that A.
africanus ate things like seeds, roots, and tubers of grasses and sedges and may have
eaten animals that feed on C4 foods. This does not necessarily mean that A. africanus
hunted vertebrate prey. They could have scavenged for meat or fed on termites, birds’
eggs, grubs, or honey. In contrast, the diet of chimpanzees, even those living in rela-
tively open and arid habitats, is derived almost exclusively from C3 foods, and their
preferred prey feed on C3 foods as well.
As we will discuss more fully later, these results are important because they tell us
australopith infants did not have as long a period of dependency as human children
do. Many anthropologists believe that a number of the fundamental features of human
foraging societies, such as the establishment of home bases, sexual division of labor,
and extensive food sharing, were necessitated by a long period of infant dependency.
If australopith infants matured quickly, then it seems likely that these features were not
yet part of the hominin adaptation.
A. GARHI
In 1999, a research team led by Berhane Asfaw and Tim White announced the dis-
covery of a new species, Australopithecus garhi, from the Awash valley of Ethiopia.
Garhi means “surprise” in the Afar language. In 1996, Asfaw, White, and their col-
leagues had recovered hominin remains from a site called Bouri, the ancient site of a
shallow freshwater lake, not far from where A. afarensis was found. The findings
included a number of postcranial bones and the partial skeleton of one individual.
These specimens are well preserved and securely dated to 2.5 mya, but they don’t con-
tain diagnostic features that can be used to assign them to a specific species. In 1997,
however, the research team discovered a number of cranial remains about 300 m
(1,000 ft) away from the original site. These pieces of the skull, maxilla (upper jaw),
and teeth come from the same stratigraphic level as the postcranial remains, and they
reveal more about the taxonomic identity of the Bouri hominins.
These creatures had small brains (approximately 450 cc), like A. afarensis and A.
africanus, perched above a very prognathic face (Figure 10.22). The canines, premo-
lars, and molars of A. garhi were generally larger than those of A. afarensis and A.
africanus. Certain detailed features of the dentition differed from A. afarensis as well.
This creature also had a sagittal crest, a fin of bone that runs along the centerline of
the skull, making it look a bit like a punk rocker’s Mohawk haircut. The Bouri spec-
imens were assigned to a new species of the genus Australopithecus because they dif-
fer from any members of the genus but lack the derived characters associated with
other hominins.
So far, the postcranial remains found near the type specimen of A. garhi cannot
be assigned to a particular species, but they do reveal some interesting developments
Sagittal crest
in the hominin lineage. Reconstructions of the humerus, radius, ulna, and femur of the
hominins at Bouri suggest that these creatures’ femurs had become longer in relation
to the humerus than in A. afarensis, while the relationship between the bones of the
upper and lower arm remained the same as in A. afarensis. The postcranial remains
suggest that there was considerable variation in the size and robustness of the Bouri
hominins. This variation may reflect the fact that males were larger than females, but
there is not yet enough evidence to be certain that this was the case.
Paranthropus
Zygomatic
arches
(b)
(a)
In the late 1930s, a retired Scottish physician and avid paleontologist named
Robert Broom discovered fossils at Kromdraai that seemed very different from A.
africanus specimens previously found at the nearby site of Sterkfontein (Figure 10.26).
The creatures at Kromdraai were much more robust (with more massive skulls and
larger teeth) than their neighbors, so Broom called his find Paranthropus robustus.
Many of his colleagues questioned this choice. They thought he had discovered
another species of Australopithecus, and so for a time most paleontologists classified
Broom’s fossils as Australopithecus robustus. Recently, how-
ever, the consensus has been that the robust early australop-
iths are so different from the gracile (more lightly built)
australopiths like A. afarensis, they should be placed in a sep-
arate genus, and Broom’s original classification has been
restored.
There is now a large sample of P. robustus fossils from
Kromdraai and the nearby site Swartkrans. These creatures
appeared about 1.8 mya and disappeared about 1 mya.
Their brains averaged about 530 cc. Paranthropus robustus
males stood about 1.3 m (41⁄2 ft) tall and weighed about 40
kg (88 lb); females were about 1.1 m (31⁄2 ft) tall and
(a) Paranthropus aethiopicus (b) Homo sapiens weighed 32 kg (70 lb). Their postcranial anatomy shows that
In these two skulls— of (a) Paranthropus they were undoubtedly bipedal. They share many of the
FIGURE 10.25 same derived features with humans that A. africanus does. In
aethiopicus and (b) a modern human—the
area of attachment of the temporalis muscle is shown in color. addition, they share with A. africanus several specialized fea-
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 267
tures for heavy chewing, but these feature are much more
pronounced in P. robustus.
It is unclear what the robust australopiths were doing
with this massive chewing apparatus. Until recently, most
anthropologists believed that P. robustus relied more on
plant materials that required heavy chewing than A.
africanus did. In other animals, large grinding teeth are
often associated with a diet of tough plant materials, and
omnivorous animals typically have relatively large canines
and incisors. Moreover, the wear patterns on P. robustus
teeth suggest that they ate very hard foods like seeds or nuts.
The carbon isotope levels in the tooth enamel of P. robustus
at Swartkrans are very similar to the values for A. africanus,
suggesting that the robust australopiths might have had
considerable overlap in their diets. Paranthropus robustus
obtained nearly a third of its diet from C4 foods, such as
grasses, roots, and tubers, or consumed animals that fed on
these kinds of foods.
A nearly complete skull of Paranthropus
FIGURE 10.26
Paranthropus boisei was a robust robustus. robustus was found at Sterkfontein in
South Africa in 1999. This species has very large jaws and molars and
Another robust australopith with large molars was dis- relatively small incisors and canines.
covered by Mary Leakey at Olduvai Gorge, Tanzania, in 1959
(Figure 10.27). This specimen, officially labeled Olduvai Hominin 5 (OH 5), was first
classified as Zinjanthropus boisei. Zinj derives from an Arabic word for “East Africa,”
and boisei comes from Charles Boise, who was funding Leakey’s research at the time.
Leakey’s find was later reclassified Paranthropus boisei because of its affinities to the
South African forms of P. robustus. (Those who advocate classifying P. robustus as Aus-
tralopithecus robustus would assign Leakey’s find to the same genus and call it A. boi-
sei.) The discovery of OH 5 was important, partly because it ended nearly 30 frustrating
years of work in Olduvai Gorge in which there had been no dramatic hominin finds. It
was only the first of a very remarkable set of fossil discoveries at Olduvai.
Essentially, P. boisei is an even more robust P. robustus—that is, a hyperrobust
australopith. Its body is somewhat larger than the body of P. robustus, and its molars
are larger than those of P. robustus, even when the difference
in body size is taken into account. The enamel is extremely
thick, and the skull is even more specialized for heavy
chewing.
In other primates, the proportion of leaves in the diet is
inversely related to the ratio of the size of first and third
molars. In P. boisei, this ratio is quite low, suggesting that
their diet consisted of leaves or seeds. However, other ele-
ments of their teeth, including rounded molar cusps, suggest
that they would not have processed leaves very efficiently.
Instead, their diet might have consisted of things like seeds,
tubers, bulbs, roots, and rhizomes.
Paranthropus boisei appears in the fossil record at about Olduvai Gorge in Tanzania has been the
2.2 mya in eastern Africa. It became extinct about 1.3 mya, FIGURE 10.27
site of many important paleontological and
although the exact date of its disappearance is not well- archaeological finds. Louis and Mary Leakey worked in Olduvai for
established. more than 30 years.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 268
Kenyanthropus
Homo habilis/rudolfensis
Fossils with a larger brain and more humanlike teeth have been discovered at
many sites in East Africa.
In 1960, while working at Olduvai Gorge with his parents, renowned paleoanthro-
pologists Louis and Mary Leakey, Jonathan Leakey found pieces of a hominin jaw, cra-
nium, and hand. The Leakeys assigned this specimen, labeled Olduvai Hominin 7 (OH
7), to the genus Homo because they believed the cranial bones indicated that it had a
much larger brain than Australopithecus had. In addition, the teeth are smaller and
have thinner enamel, and the dental arcade is more parabolic than in the australop-
iths. The skulls are more rounded, there are fewer air pockets in the bottom of the
skull, the face is smaller and protrudes less, and the jaw muscles are reduced in size
compared with the australopiths. Louis Leakey named the species Homo habilis, or
“handy man,” because he believed he had found the hominin responsible for the sim-
ple, flaked-stone tools discovered nearby. Later, the Leakeys found several more fos-
sils that they assigned to H. habilis, including pieces of a second cranium, part of a
foot, and more teeth. These fossils range in age from 1.6 to 1.9 million years old.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 269
0 5 cm
dental enamel indicates that specimens assigned to the species habilis and rudolfensis
had rapid apelike developmental patterns, like other australopiths.
HOMININ PHYLOGENIES
It is difficult to infer the phylogenetic relationships among the Plio-Pleistocene
hominins.
among all of the early hominin species that we have described Later Homo
0
here. Unfortunately, as the hominin fossil record has become Homo
richer, this task has become harder. The problem is illustrated by erectus
the phylogeny shown in Figure 10.32, which accompanied the
first published description of Kenyanthropus platyops. Clearly, a
Paranthropus
vast number of different phylogenies could be constructed, boisei
1
depending on which relationships are chosen and which are Homo Paranthropus
rejected. ergaster robustus
The problem arises from the extensive convergence and par-
Australopithecus
allelism in hominin evolution. This means that many alternative (Homo) habilis
phylogenies are equally plausible, depending on what is assumed ? Australopithecus
?
? africanus
to be homologous and what is assumed to be convergent. Ran- 2
that the branching pattern of the best tree depends on which Australopithecus
groups of traits are included. For example, the best fit based on anamensis
Homo
0
Paranthropus
Paranthropus
boisei
robustus
1
Australopithecus
Paranthropus
aethiopicus
afarensis
3
Australopithecus
africanus
4
(a)
Homo
Paranthropus
0
boisei
Paranthropus
robustus
1
Millions of years ago
Australopithecus
africanus
traits other than those involved in heavy chewing are homologous, then we might
obtain the phylogeny shown here, in which Australopithecus afarensis gives rise to
Australopithecus africanus, who in turn gives rise to Homo species and the later 4
paranthropines. According to this phylogeny, the similarities between Paranthropus
aethiopicus and the other paranthropines are due to parallel evolution. (b)
of creature that linked the apes of the Miocene to the earliest members of our own
genus, Homo.
In the discussion that follows, we first consider why bipedal locomotion evolved,
and we then review what we can infer about the behavior and life history of these
hominins based on our knowlwdge of the behavioral ecology of other primates, par-
ticularly chimpanzees.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 273
More wind
Bipedal loco- Lower
FIGURE 10.34 temperatures
motion helps
an animal living in warm climates to keep
cool by reducing the amount of sunlight
that falls on the body, by increasing the
animal’s exposure to air movements, and Higher
by immersing the animal in lower- temperatures Less wind
temperature air.
heat stress and lessens the amount of water necessary for evaporative cooling, in three
different ways (Figure 10.34):
1. The sun strikes a much smaller fraction of the body of an erect hominin than of
a quadrupedal animal of the same size. An erect hominin absorbs sunlight
directly only on the top of its head and shoulders, while a quadruped absorbs
the energy of the sun across its whole back and head. (Think about how much
harder it is to get a tan standing up than lying down.) This difference is impor-
tant because near the equator the sun is overhead during much of the middle of
the day.
2. The air is warmer near the ground than it is higher up. The sun heats the soil,
and the soil radiates and conducts heat back to the air, so air temperatures are
highest close to the ground.
3. Wind velocities are higher 2 m (about 6 ft) off the ground than they are 1 m
(about 3 ft) off the ground. Because moving air increases the efficiency of evap-
orative cooling, an erect hominin is better able to withstand the heat of the day.
Using standard engineering methods, Wheeler calculated that these three factors would
reduce heat stress and water consumption in bipedal hominins compared with
quadrupeds. Interestingly, the same calculations show that loss of body hair provides
a thermoregulatory advantage to a bipedal creature, but not to a quadruped of the
same size.
Bipedal locomotion leaves the hands free to carry things. The ability to carry
things is, in a word, handy. Quadrupeds can’t carry things in their hands without inter-
fering with their ability to walk and to climb. As a consequence, they must carry things
in their mouths. Some Old World monkeys pack great quantities of food into their
cheeks to be chewed and swallowed later. Other primates must eat their food where
they find it—a necessity that can cause problems when food is located in a dangerous
place, or when there is a lot of competition over food. Bipedal hominins can carry
larger quantities of food in their hands and arms.
Bipedal posture allows efficient harvesting of fruit from small trees. Indiana Uni-
versity anthropologist Kevin Hunt has argued that the anatomy of A. afarensis is well
suited to standing erect but not well designed for efficient bipedal walking. Hunt thinks
that bipedal posture was favored because it allows efficient harvesting of fruit from the
small trees that predominate in African woodlands. Two kinds of data support this
hypothesis. First, Hunt found that chimpanzees rarely walk bipedally, but they spend
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 275
a lot of time standing bipedally as they harvest fruit from small trees (Figure 10.35).
Using their hands for balance, they pick the fruit and slowly shuffle from depleted
patches to fresh ones. Standing upright allows the chimpanzees to use both hands to
gather fruit, and the slow bipedal shuffling allows them to move from one fruit patch
to another without lowering and raising their body weight. Second, Hunt argues that
many anatomical features of A. afarensis are consistent with standing upright but not
with walking bipedally. He points out that the very broad pelvis of A. afarensis would
make a stable platform for standing on both feet. Hunt also argues that the features
of the shoulders, hands, and feet that many anthropologists believe are adaptations for
climbing trees are actually adaptations that enabled A. afarensis to hang by one hand
while standing bipedally and feeding with the other hand.
Any or all of these hypotheses may be correct. Bipedalism might have been favored
by selection because it was more efficient than knuckle walking, because it allowed
early hominins to keep cool, because it enabled them to carry food or tools from place
to place, and/or because it enabled them to feed more efficiently. And once bipedalism
had evolved, it might have facilitated other forms of behavior, such as tool use.
We can use what we know about primate behavior and ecology to make some
inferences about the foraging behavior and social organization of early hominins.
The creatures that left the fossils we have described in this chapter probably spent their
days (and nights) much like other primates do: moving around their home ranges look-
ing for food and water, caring for their offspring, looking out for dangerous predators,
monitoring mating opportunities, and so on. We can use what we have learned about Chimpanzees
FIGURE 10.35
primate behavior and ecology to develop some tentative speculations about the sometimes
behavioral strategies of early hominins. stand bipedally as they harvest fruit
from small trees. They use one hand
for balance and feed with the other,
Increased seasonality in rainfall favors a greater dependence on foods available shuffling slowly from one food patch
during the dry season: corms, tubers, and meat. to another.
Herbivorous animals that live in seasonal environments are faced with quite a variable
food supply. When the rain falls, grasses sprout, flowers blossom, fruit ripens, and
there are many pools of freshwater (Figure 10.36a). Food is abundant and diverse.
Because food is plentiful, there is little competition among species over access to food.
During the dry season, grasses go to seed, flowers die, and all but a few permanent
water holes dry up (Figure 10.36b). Both the abundance and diversity of food are
In seasonal
FIGURE 10.36
habitats, the
landscape changes dramatically over the
course of the year. Here you see the
same area in Amboseli just after the
seasonal rains have ended (a) and at the
end of the long dry season (b).
(a) (b)
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 276
In seasonal
FIGURE 10.37
habitats,
resource availability is strongly affected
by rainfall. (a) After the rains, lush
growth sprouts, and baboon foods are
abundant and diverse. (b) By the end of
the dry season, the ground is parched,
new growth has withered, and less food
is available to baboons.
(a) (b)
greatly reduced (Figure 10.37). Because there is very little to eat for most herbivorous
animals, intense competition erupts among certain species. Many species respond to
this problem by focusing their foraging efforts on a particular type of food during the
dry season. In Amboseli National Park, for example, baboons spend much of their
time during the dry season digging up grass corms (the bulbous storage organs in grass
stems). Another way to cope with shortages of plant foods is to become more omniv-
orous (Figure 10.38). Baboons at many different sites increase their meat intake dur-
ing the dry season (Figure 10.39).
Like baboons, some of the early hominins may have adapted to increased sea-
sonality by altering their feeding habits. Using methods that enable them to track
short-term changes in the composition of dental enamel in individuals, Sponheimer
and his colleagues have discovered that P. robustus’s diet varied considerably across
seasons and across years. Similarly, close inspection of the wear patterns on the teeth
A male of A. africanus suggest that their diets also varied seasonally. In contrast, A. afaren-
FIGURE 10.38
baboon con- sis, which was largely confined to wooded areas, shows little sign of such dietary
sumes a newborn bushbuck in the Oka-
shifts.
vango Delta of Botswana.
0.6
Wet
Dry
Predation attempts per animal per month
0.4
0.2
Chimpanzees
FIGURE 10.40
hunt small
mammals, particularly red colobus mon-
keys. (a) A red colobus monkey carry-
ing an infant leaps from one tree to
another at Gombe Stream National
Park in Tanzania. (b) A male chim-
panzee at Gombe consumes a red
colobus monkey. (Photographs courtesy
of Craig Stanford.)
(a) (b)
Chimpanzees are effective hunters who capture monkeys, bushpigs, and small
antelope, but they rarely scavenge.
In the past, many anthropologists thought it was unlikely that the early hominins
ate very much meat. Hunting, they argued, requires careful planning, subtle coordi-
nation, and extensive reciprocity. These interactions would require cognitive and lin-
guistic skills that the relatively small-brained australopiths must have lacked. Studies
of chimpanzee hunting by Christophe and Hedwige Boesch-Achermann in the Taï For-
est of Ivory Coast, Craig Stanford at Gombe Stream National Park in Tanzania, and
John Mitani and David Watts in Ngogo, Uganda, suggest that we must revise this view.
At Taï, Gombe, and Ngogo, chimpanzees participate in hunts approximately once
every two or three days. Most predatory activity focuses on monkeys, particularly red
colobus (Figure 10.40). Gombe chimpanzees also prey on bushpigs and bushbucks
(sheep-sized forest antelope) that they encounter on the forest floor (Figure 10.41).
Unlike some other meat-eating species, chimpanzees never feed on the carcasses of
dead animals that they encounter opportunistically, and they rarely take fresh kills
from other predator species.
At Gombe, Ngogo, and Taï, many red colobus hunts begin when the chimpanzees
encounter a group of monkeys in the forest or detect them at a distance. They often
spend some time on the ground gazing up into the canopy before they begin to hunt.
Then at least one chimpanzee climbs into the trees and gives chase. Others watch from
the ground.
Although most hunts at Gombe and Ngogo seem to arise out of opportunistic
encounters with groups of monkeys, about half of the hunts at Taï begin before there
Other
Bushbuck
Colobus
Bushpig
is any sign that monkeys are nearby. In these cases, the chimpanzees fall completely
silent, stay very close to one another, frequently change direction, and stop often to
search for monkeys in the canopy. The chimpanzees in the three study areas are effec-
tive hunters, making kills in approximately half of their hunts.
These observations invalidate the argument that a small-brained hominin must
have lacked the necessary cognitive and linguistic skills for hunting. However, they do
not tell us how likely it is that the early hominins actually hunted.
Early hominins may have used tools like contemporary chimpanzees do.
As we will see in the next chapter, the earliest identifiable tools first appear about
2.5 mya, and it is possible that one of the early hominins discussed here manufactured
these tools. However, the absence of identifiable tools does not mean that early
hominins were not tool users.
Chimpanzees make and use tools to perform a variety of different tasks. They use
long, thin branches, vines, stems, sticks, and twigs to poke into ant nests, termite
mounds, and bees’ nests to extract insects and honey. They sometimes scrape marrow
and other tissue from the bones and braincases of mammalian prey with wooden
twigs. They pound stone hammers against heavy flat stones, exposed rocks, and roots
to crack open hard-shelled nuts (Figure 10.42). They wad up leaves, dip them into
pools of rainwater that have collected in hollow tree trunks, and then suck water from
the leafy “sponges.”
Moreover, chimpanzees modify natural objects for specific purposes. That is,
In West
FIGURE 10.42 they are toolmakers, as well as tool users. For example, chimpanzees at Mount
Africa, chim-
Assirik in Senegal, studied by Caroline Tutin of the University of Stirling and
panzees use stones to crack open hard-
shelled nuts. William McGrew of the Miami University, Ohio, use twigs to extract termites from
their mounds. The twigs are first detached from the bush or shrub; then leaves grow-
ing from the stem are stripped off, the bark is peeled back from the stem, and the twig
is clipped to an appropriate length. Chimpanzees do not modify all of the materials
that they use as tools. The stone hammers and anvils that chimpanzees use to crack
open hard-shelled nuts are carefully selected, used repeatedly, and sometimes moved
from one site in the forest to another, but they are not deliberately altered by the
chimpanzees. However, none of these tools would leave a trace in the archaeologi-
cal record.
Approximately half of the tasks for which chimpanzees manufacture and use tools
are related to food processing and food acquisition. In some cases, these tools allow
chimpanzees to exploit food resources that other animals cannot use. For example,
many animals are attracted to termite mounds when the winged insects emerge in mas-
sive swarms to form new colonies, but only chimpanzees can obtain access to the ter-
mites inside the mounds at other times of the year. Some of the shells of nuts that
chimpanzees eat in the Taï Forest are so hard that they can be opened only by repeated
battering of the shell with a stone hammer.
Despite the examples of tool use by chimpanzees, humans are the only primates
to use tools extensively. William McGrew argues that the reason chimpanzees don’t
make greater use of tools is that they can’t transport them easily (Figure 10.43). If a
chimpanzee can’t carry a tool from place to place, then it must manufacture its tools
at the site where they will be used. As a consequence, it doesn’t pay for chimpanzees
to put a lot of effort into making tools, so the tools they use are crude.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 279
25
Large prey
20
Small prey
Number of kills
15
10
In the Taï Forest,
FIGURE 10.44
5 chimpanzees some-
times share their kills. Infant and juvenile mon-
0 keys (small prey) are generally consumed by the
Keeps all Loses all by Gives up all Splits prey captor, but adult monkey carcasses are generally
force
divided by the captor and shared with other
Distribution of prey by owner chimpanzees.
10_Hominoid to Hominin [4p]:How Humans Evolved [5e] 10/10/08 12:08 PM Page 280
FURTHER READING
STUDY QUESTIONS
KEY TERMS
hominins tibia
torque humerus
abductors endocranial volume
ilium subnasal prognathism
femur diastema
foramen magnum sagittal crest
pneumatized temporalis muscle
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 282
Chapter 11
The Oldowan Toolmakers
Complex Foraging Shapes
Human Life History
Evidence for Complex Foraging
by Oldowan Toolmakers
Archaeological Evidence for Meat Eating
Hunters or Scavengers?
Domestic Lives of Oldowan Toolmakers
Back to the Future: The Transition to
Modern Human Life Histories
H
ominins have been making stone tools for several million years. The first
identifiable stone tools were found in the Awash region of Ethiopia and are
dated to about 2.5 mya. Similar tools found at a number of other sites in
East Africa are between 1.7 and 2.4 million years old. In South Africa, the earliest
stone tools date to about 2 mya. Tools made from bone appear in the fossil record
between 2 and 1 mya at Olduvai Gorge and in South Africa.
282
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 283
The earliest toolmakers were already quite proficient at their craft. At a West
Turkana site that dates to 2.3 mya, researchers have found an extensive array of stone
artifacts, including flakes (small, sharp chips), cores, hammer stones, and debris from
manufacturing. Preservation conditions were so good that workers were able to fit a
number of flakes back onto the cores from which they had been struck. This recon-
struction revealed that early hominins removed as many as 30 flakes from a single
core, maintaining precise flaking angles during the entire toolmaking sequence.
These artifacts, collectively referred to as the Oldowan tool industry, are very sim-
ple. They consist of rounded stones, like the cobbles once used to pave city streets, that
have been flaked (chipped) a few times to produce an edge (Figure 11.1). Toolmaking
(knapping) leaves telltale traces on the cobble cores, so it’s possible to distinguish nat-
ural breakage from deliberate modification. The Oldowan artifacts are quite variable
in their shape and size, but this variation does not seem to be related to differences in
how the tools were used, how they were made, or how toolmakers thought their tools
should look. Instead, the tools vary because they were made out of different raw mate-
rials. There is now evidence that the flakes struck from these cobbles were at least as
useful as the cores themselves (Box 11.1).
Bifacial chopper
Hammer stone
Flake scraper
Discoid
Flake
Box 11.1
Ancient Toolmaking
and Tool Use
Recently, Kathy Schick and Nicholas Toth of Indiana Uni- can be used for an impressive variety of tasks, even for
versity have done many experiments with simple stone butchering large animals like elephants. In contrast,
tools. They have mastered the skills needed to manufacture although the cores can be used for some tasks, such as chop-
the kinds of artifacts found at Oldowan sites and learned ping down a tree to make a digging stick or a spear, or
how to use them effectively. Their experiments have pro- cracking bones to extract marrow, they are generally much
duced several interesting results. Schick and Toth believe the less useful. Schick and Toth’s conclusion is supported by
Oldowan artifacts that archaeologists have painstakingly microscopic analysis of the edges of a small number of
collected and described are not really tools at all. They are Oldowan flakes, which indicates that they were used for
cores left over after striking off small, sharp flakes, and the both woodworking and butchery.
flakes are the real tools. Schick and Toth find that the flakes Schick and Toth have also been able to explain the func-
1 2 3
284
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 285
tion of enigmatic objects that archaeologists call spheroids. Perhaps the most remarkable conclusion that Schick
These are smooth, approximately spherical pieces of quartz and Toth drew from their experiments is that early
about the size of a baseball. Several suggestions for their hominin toolmakers were right-handed. Schick and Toth
function were put forth, including processing plants and found that right-handers usually hold the hammer stone in
bashing bones to extract marrow. Some researchers thought their right hand and hold the stone to be flaked in the left
the spheroids were part of a bola, a hunting tool used on the hand. After driving off the first flake, they rotate the stone
grasslands of Argentina. In a bola, three stones are connected clockwise and drive off the second flake. This sequence
by leather thongs and thrown so that they tangle the prey’s produces flakes in which the cortex (the rough, unknapped
legs. Schick and Toth have shown there is a much more plau- surface of the stone) is typically on the right side of the
sible explanation for these stones. If a piece of quartz is used flakes (Figure 11.2). On flakes made by left-handers, the
as a hammer to produce flakes, bits of the hammer stone are cortex is typically on the left side. With this result in hand,
inadvertently knocked off. The hammer surface is no longer Schick and Toth studied flakes from sites at Koobi Fora,
flat, so the hammer stone is shifted in the toolmaker’s hand. Kenya, dated to 1.9 to 1.5 mya. Their results suggest that
Thus the hammer gradually becomes more and more spher- most of the individuals that made these flakes were right-
ical. After a while, a quartz hammer becomes a spheroid. handed.
manufacturing tools so that we can compare industries from different times and places.
The late J. Desmond Clark of the University of California, Berkeley devised a scheme
for classifying modes of production that we will use here. According to this scheme,
crude flaked pebble tools like those associated with the Oldowan tool industry are clas-
sified as Mode 1 technology. The distinction between industry and mode is not very
meaningful for our discussion of early hominins, who used simple techniques to cre-
ate a very limited set of tools, but it will become significant later when there are major
regional and temporal variations in the composition of tool kits and modes of
production.
We do not know which hominin species were responsible for making the tools.
As you learned in Chapter 10, a sizable number of hominin species were running
around in East Africa between 2.5 and 1.7 mya, and it is not clear which of these
species made the Oldowan tools. Australopithecus garhi is probably the best candidate
for “first engineer.” At Bouri, animal bones bearing the distinctive marks made by
stone tools are found in the same strata as the fossils of A. garhi. Although no tools
were found with the fossils at Bouri, many tools have been found at the nearby site of
Gona, which is dated to 2.5 mya. Thus, A. garhi may have been the first stone tool-
maker. But remember that A. rudolfensis, Kenyanthropus platyops, and Paranthropus
aethiopicus were probably all present in East Africa at the same time; they could have
made the tools found at Gona but impolitely failed to leave their own fossils behind.
It is also possible that the earliest stone tools were made by species that do not
appear until later in the fossil record. Stone tools are more durable than bones, so the
archaeological record is usually more complete than the fossil record. This means that
the earliest tools typically appear in the fossil record before the first fossil of the crea-
ture that made them (see Box 9.3 for more discussion of this topic). Thus, the first
stone toolmaker may have been an australopith, Homo habilis, or even Homo
ergaster, the oldest member of our own genus, who will be discussed in more detail in
Chapter 12. Because of this ambiguity, in the rest of this chapter we will refer to the
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 286
hominins who made the tools of this period as Oldowan hominins or Oldowan
toolmakers.
Homo ergaster appeared in Africa about 1.8 mya. Unlike the situation with H.
habilis or A. rudolfensis, there is little doubt that these creatures belong in the same
genus as Homo sapiens. As you will learn in the next chapter, Homo ergaster was a
fully terrestrial creature with a large body—one quite similar to our own. These crea-
tures developed more slowly than early hominins, and males and females were about
as different in size as men and women are today. Thus, about 2 mya selection must
have favored these more humanlike characteristics in one of the early hominin line-
ages. That we don’t know which early hominin is ancestral to humans is annoying but
not really fatal, because it seems very likely that whoever made the Oldowan tools is
the culprit. By combining knowledge of contemporary foraging peoples with careful
study of Oldowan tools, the sites where these tools were found, and the marks that
they make on animal bones, we can learn a lot about the transition from a bipedal ape
to a creature much more like modern humans.
In the discussion that follows, we will first see how a reliance on highly produc-
tive but very hard-to-learn foraging skills distinguishes humans from other primates,
and how this novel foraging niche may have led to the evolution of other novel fea-
tures of the human life cycle, such as slow maturation and reduced sexual dimorphism.
We will then consider the archaeological evidence from Oldowan sites indicating that
these hominins had begun to shift to a subsistence economy based on more challeng-
ing foraging techniques.
Hillard Kaplan and Jane Lancaster of the University of New Mexico, along with Kim
Hill and A. Madgalena Hurtado of Arizona State University, recently argued that the
evolution of modern human life history was driven by a shift to valuable but hard-to-
acquire food resources. They rank food resources into three categories according to
the difficulty of acquisition:
1. Collected foods can be simply collected from the environment and eaten. Exam-
ples include ripe fruit and leaves.
2. Extracted foods come from things that don’t move but are protected in some
way. These things must be processed before the food can be eaten. Examples
include fruits in hard shells, tubers or termites that are buried deep underground,
honey hidden in hives high in trees, and plants containing toxins that must be
extracted before the plants can be eaten.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 287
80 80
(a) Chim-
Percentage of diet
FIGURE 11.3
60 60 panzees spend
most of their time feeding on collected
40 40 foods like fruit and leaves, which can be
eaten without processing. (b) Human
foragers get most of their calories from
20 20
extracted foods like tubers, which must
be processed before they can be eaten,
0 0
Collected Extracted Hunted Collected Extracted Hunted and hunted foods like animal game,
Method of food acquisition Method of food acquisition which must be caught or trapped.
(a) (b)
3. Hunted foods come from things that run away and thus must be caught or
trapped. They may also need to be extracted and processed before consumption.
Vertebrate prey are the prime example of hunted foods for both humans and
chimpanzees.
Apes, especially chimpanzees, are the “brainiacs” of the primate world, at least
when it comes to foraging. Gorillas and orangutans use elaborate routines to process
some plant foods, but they don’t hunt. Chimpanzees have a broader diet, including col-
lected, hunted, and extracted foods, the latter two requiring considerable skill to
acquire and process. Both orangutans and chimpanzees use tools to process some kinds
of foods. Moreover, as we will see in Chapter 15, different ape populations use dif-
ferent techniques for the acquisition of extracted foods like hard-shelled nuts. How-
ever, even these clever apes do not come close to human expertise in foraging.
Kaplan and his colleagues note that contemporary foraging peoples depend on
extracted and hunted foods to a much greater extent than chimpanzees do. Figure 11.3
compares the average dependence of chimpanzees and humans on each of the food
types: collected, extracted, and hunted. The general pattern is clear: chimpanzees are
overwhelmingly dependent on collected resources, but human foragers get almost all
of their calories from extracted or hunted resources.
Unlike other predators, humans must learn a very diverse set of hunting skills.
Most large mammalian predators capture a relatively small range of prey species using
one of two methods: they wait in ambush, or they combine a stealthy approach with
fast pursuit. Once the prey is captured, they process it with tooth and claw. In con-
trast, human hunters use a vast number of methods to capture and process a huge
range of prey species. For example, the Aché, a group of foragers who live in
Paraguay, have been observed to take 78 different species of mammals, 21 species of
reptiles, 14 species of fish, and over 150 species of birds using a dizzying array of tech-
Meat makes up
niques that depend on the prey, the season, the weather, and many other factors (Fig- FIGURE 11.4
about 70% of
ure 11.4). Some animals they track—a difficult skill that entails a great deal of the diet of the Aché, a group of for-
ecological and environmental knowledge (see Chapter 14). Other animals they call by agers from Paraguay. Here an Aché man
imitating the prey’s mating or distress sounds. Still other animals they trap with snares takes aim at a monkey. (Photograph
or traps, or smoke out of burrows. They capture and kill animals using their hands, courtesy of Kim Hill.)
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 288
arrows, clubs, or spears. And this is just the Aché; if we included the full range of
human habitats, the list would be immeasurably longer.
It takes a long time to learn this range of skills. Among the Aché, men’s hunting
efficiency peaks at about age 35. Twenty-year-old men manage to capture only about
a fourth of the maximum. Kaplan and Hill made strenuous efforts to become com-
petent hunters when they were living with the Aché, but they could not come close to
the production of 20-year-old Aché men.
Efficient extraction of resources also requires considerable skill. Nicholas Blurton
Jones, a scientist at the University of California, Los Angeles who has studied the
Hadza and the !Kung, two foraging groups, describes digging up deeply buried
tubers from rocky soil as a complex mining operation involving much clever engi-
neering of braces and levers (Figure 11.5). Among the Hiwi, a group of foragers who
live in the tropical savanna of Venezuela, women do not achieve maximum efficiency
in root acquisition until they are between 35 and 45 years old. Ten-year-old girls get
only 10% as much as older, highly skilled women. Among the Aché, rates of starch
A !Kung San extraction from palms and honey extraction also peak when people are in their 20s.
FIGURE 11.5
woman carries
her young child on her back and digs for
roots and tubers in the Kalahari Desert
A reliance on hunting and extractive foraging favors food sharing and division
of Botswana. (Photograph courtesy of of labor in contemporary foraging groups.
Nicholas Blurton Jones.)
In all contemporary foraging groups, hunting and extractive foraging are associ-
ated with extensive food sharing and sexual division of labor. In nearly all foraging
groups, men take primary responsibility for hunting large game, and women take pri-
mary responsibility for extractive foraging (Figure 11.6). This division of labor makes
sense on two grounds: First, hard-to-learn techniques reward specialization. It takes
a long time to learn how to be a good hunter, and it takes a long time to learn how to
dig tubers. This means that everyone is better off if some individuals specialize in hunt-
ing and others specialize in extractive foraging. Second, because child care is more
compatible with gathering than with hunting, and lactation commits women to child
care for a substantial portion of their adult lives, it makes sense that men specialize in
4,000
Men
Women
Number of calories produced per day
3,000
2,000
1,000
Data on foraging behavior for three
FIGURE 11.6
well-studied modern foraging groups:
the Aché, the Hadza, and the Hiwi. Men and women in these
groups specialize in different foraging tasks. Men hunt, and 0
women concentrate on extractive foraging. Meat Roots Fruits Other
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 289
hunting and women specialize in extractive foraging. Of course, this all works only if
members of the group regularly share food.
However, a reliance on meat eating favors the evolution of food sharing. Hunt-
ing is an uncertain endeavor. The best hunters can come home empty-handed; and even
if hunters caught enough each day to feed themselves, a week or two of bad luck could
lead to starvation. If several hunters share their catch, however, the chance of starva-
tion is much lower (Box 11.2).
In foraging societies, food sharing and division of labor lead to extensive flows
of food between people of different ages and sexes.
Remember from Chapter 10 that the total amount of food shared by chimpanzees
is small: mothers share with their offspring, and males share small amounts of meat,
mainly with other adult males. This means that, once they are weaned, chimpanzees
obtain virtually all of their own food themselves. The story for other primates is very
similar. Thus, it seems likely that such self-sufficiency after weaning is the ancestral
state in the hominin lineage.
The economy of human foragers is strikingly different: some people produce much
more food than they consume, and others consume much more than they produce.
Over the last 20 years, anthropologists have done careful quantitative studies of the
subsistence economies of a number of different foraging groups. In these societies,
anthropologists observed people’s daily behavior, measuring how much food they pro-
duced and how much they consumed. For three of these groups—the Aché, the Hiwi,
and the Hadza—researchers have meticulously computed average food production
and consumption for men and women of different ages. Kaplan and his colleagues
compiled these data to compare patterns of food production across societies. Their
analysis reveals striking differences in the foraging economy of humans and chim-
panzees, and important changes in productivity over the life course.
Figure 11.7 shows that human young continue to depend on others for food long
after they are weaned. Men become self-sufficient around the age of 17, and women
do not produce enough to feed themselves until they are in their late 40s. Older men
8 8
Men’s
Number of kcal/day (× 1,000)
production
Number of kcal/day (× 1,000)
6 6
4 Men’s 4 Women’s
consumption Women’s
consumption production
2 2
0 0
10 20 30 40 50 60 70 10 20 30 40 50 60 70
Age Age
(a) (b)
Data from three contemporary foraging groups show that both (a) men and
FIGURE 11.7
(b) women do not become self-sufficient in terms of food production until
they are adults. Adult men produce many more calories than they consume; adult women are approxi-
mately in balance.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 290
Box 11.2
Why Meat Eating
Favors Food Sharing
Many anthropologists believe that a heavy dependence on hunting groups contained 7 males on average. Dividing 0.84
meat makes food sharing necessary. Let’s examine how by 7 gives the average number of monkeys killed per male
food sharing provides insurance against the risks inherent in per hunt, which comes out to 0.12. Thus, on any given day,
hunting. Hunting, especially for hunters who concentrate on each male chimpanzee had only a 12% chance of making a
large game, is a boom-or-bust activity. When a hunter kill, which is less than the 20% chance we posited for the
makes a kill, a lot of very high-quality food becomes avail- five human hunters at the outset of our example.
able. Hunters are often unlucky, however, and each time one Now let’s consider how sharing food alters the proba-
sets out to hunt there is a fairly high probability of return- bility of starvation for our human hunters. If each hunter
ing empty-handed and hungry. Food sharing greatly reduces has a 0.8 chance of coming back empty-handed, then the
the risks associated with hunting by averaging returns over chance that all five hunters will come back on a given
a number of hunters. evening without food is
To see why this argument has such force, consider the
following simple hypothetical example. Suppose there are 0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ⬇ 0.33
five hunters in a group that subsists entirely on meat.
Hunters are able to hunt every day, and each hunter has a Thus, on each day there is a 1-in-3 chance that no one will
1-in-5 (0.2) chance of making a kill and a 4-in-5 (0.8) make a kill. If the kill is large enough to feed all members
chance of bringing back nothing. Further, suppose that peo- of the group, then no one will go hungry as long as some-
ple starve after 10 days without food. We can calculate the one succeeds. The chance that all five hunters will face star-
probability of starvation for each hunter over any 10-day vation during any 10-day period is
period by multiplying the probability of failing on the first
day (0.8) by the probability of failing on the second day 0.33 ✕ 0 .33 ✕ 0.33 ✕ 0.33 ✕ 0.33 ✕ 0.33 ✕
(0.8), and so on, to get 0.33 ✕ 0.33 ✕ 0.33 ✕ 0.33 ⬇ 0.000015
0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ✕ Sharing reduces the chance of starvation from 1 chance
0.8 ✕ 0.8 ✕ 0.8 ✕ 0.8 ⬇ 0.1 in 10 to roughly 1 chance in 60,000. Clearly, the risks asso-
ciated with hunting could be reduced even further if there
Thus, there is about a 10% chance that a hunter will starve were alternative sources of food that unsuccessful hunters
over any 10-day period. With these odds, it is impossible for might share. For example, suppose one member of the
people to sustain themselves by hunting alone. group hunted while the others foraged, and they all con-
A comparison with chimpanzee hunting provides good tributed food to a communal pot.
reason to think that these probabilities are realistic for The fact that food sharing is mutually beneficial is not
early hominins. Craig Stanford of the University of South- enough to make it happen. As we pointed out in Chapter 7,
ern California and his colleagues have carefully analyzed food sharing is an altruistic act. Each individual will be bet-
records of hunting by chimpanzees at Gombe Stream ter off if he or she gets meat but does not share it. For shar-
National Park in Tanzania. In about half of the hunts, the ing to occur among unrelated individuals, as it often does in
chimpanzees succeeded in killing at least one monkey, and contemporary foraging societies, those who do not share
sometimes they made more than one kill. The average num- must be punished in some way, such as by being excluded
ber of monkeys killed per hunt was 0.84. However, the from future sharing or by being forced to leave the group.
290
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 291
Men Women
Nukak
Hiwi
Aché
Onge
Anbarra
Arnhem
Hadza
also depend on others for their daily needs. These deficits are made up by the pro-
duction of young and middle-aged men, and to a lesser extent, postmenopausal
women. In contrast, chimpanzees obtain very little of their food from others after they
are weaned.
Less detailed data from other foraging groups are consistent with this pattern. Fig-
ure 11.8 shows that men contribute more than half of the total calories consumed in
seven of the nine foraging groups for which the necessary data are available. Notice
that all of these groups live in tropical habitats. It seems likely from historical and
ethnographic accounts that temperate and arctic foragers depended even more on meat
than do tropical foragers, and, thus, in these societies men may make an even bigger
caloric contribution.
Selection may have favored larger brains, a prolonged juvenile period, and a
longer life span because these traits make it easier to learn complex foraging
methods.
greater intelligence. Reliance on complex learned foraging skills would also favor the
evolution of a prolonged juvenile period. As we all know, learning takes time. You
can’t become a proficient skier, baker, or computer programmer in a day; practice and
experience are needed. Similarly, learning the habits of animals and the lore of plants,
acquiring the knowledge for tracking animals and the skills for shooting a bow or
blowgun, and becoming adept at extracting starch from baobab pulp require years of
practice. Thus, it is plausible that selection favored a longer juvenile period to allow
human children the time to acquire the skills they needed.
A prolonged juvenile period generates selection for a long life span. It is often said
that time is money. In evolution, “time is fitness.” To see why, suppose that two geno-
types, A and B, have the same number of children on average, but type A completes
reproduction in 30 years and type B in 60 years. If you do a bit of math, you will see
that type A will have twice the population growth rate as type B, and it will quickly
replace type B in the population. This means that a prolonged juvenile period is costly
and will not be favored by natural selection unless it causes people to have sufficiently
more children over their life span. Human childhood is like a costly investment; it
costs time, but the added time allows learning that produces more capable adults.
Like any expensive investment, it will pay off more if it is amortized over a longer
period. (The same logic explains why you are willing to spend more on something
that you will use for a long time than on something you will use for a short time and
then discard.) Selection favors a longer life because it allows people to get more ben-
efit from the productive foraging techniques they learned during the necessary, but
costly, extended juvenile period.
Food sharing and division of labor lead to reduced competition between males
and reduced sexual dimorphism.
niques to a much greater extent than other primates do, and this shift can explain the
evolution of the main features of human life history. To link these points, we need to
consider the evidence that Oldowan hominins had begun to rely on hunting and
extractive foraging to make a living.
As described in Box 11.1, Kathy Schick and Nicholas Toth of Indiana University
have mastered the skills needed to manufacture the kinds of artifacts found at
Oldowan sites, and they have attempted to use them for different kinds of foraging
activities. Schick and Toth found that stone flakes struck from cobble cores can be used
for an impressive variety of tasks, even for butchering large animals like elephants. The
cores can be used for a more limited number of jobs, such as chopping down a tree to
make a digging stick or a spear. Microscopic analysis of the edges of a small number
of Oldowan tools indicates that they were used for both woodworking and butchery.
Wear patterns on bone tools from South Africa suggest that they were used to
excavate termite mounds.
Extractive foraging, which is commonly done with wooden digging sticks by mod-
ern peoples, is likely to leave few traces in the archaeological record. However, one
interesting piece of evidence suggests that hominins from this period were extractive
foragers. During their excavations at Swartkrans, Robert Brain and his coworkers
identified a sizable number of broken bones that had wear patterns suggesting that they
had been used as tools. Recently, Lucinda Backwell of the University of the Witwaters-
rand and Francesco d’Errico of the Institut de Préhistoire et de Géologie du Quater-
naire in France carefully analyzed these bones to find out how they were used. First
the researchers used freshly broken bones to do a number of foraging tasks, including
digging in hard soil for tubers and digging in a termite mound. Each activity creates a
distinctive wear pattern that can be detected under microscopic analysis. Then they
compared these wear patterns with those on the fossil bones found at Swartkrans.
This analysis indicated that the fossil tools were used for digging in termite
mounds. Figure 11.9 shows the wear patterns on the fossil tool and two experimen-
tal tools, one used for digging in soil and the other used for excavating a termite
Experiments
FIGURE 11.9
indicate that
bone tools found at Swartkrans were
used to excavate termite mounds. The
tool shown in (a) is a cast of the origi-
nal tool found at Swartkrans. The
experimental tools in (b) and (c) were
used for subsistence tasks, digging for
tubers and digging for termites. The
wear pattern on the Swartkrans fossils
was most similar to that on the experi-
mental tool used to dig for termites.
(a) (b) (c)
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 294
mound. All of the tools have a smooth, rounded point. However, the tool experi-
mentally used for digging in soil has deep marks of different depths going in all direc-
tions. In contrast, the tool used to dig in termite mounds has fine, parallel grooves. The
fossil-bone tools closely resemble the experimental tool used for digging in termite
mounds, so it seems likely that this is what these tools were used for. If this is correct,
then Oldowan hominins were using tools to do extractive foraging.
At several archaeological sites in East Africa, Oldowan tools have been found
along with dense concentrations of animal bones.
Archaeological sites with early stone tools occur at the Olduvai Gorge of Tanzania,
Koobi Fora in Kenya, and a number of sites in Ethiopia. The sites in Bed I of the Oldu-
vai Gorge excavated by Mary Leakey have been analyzed the most extensively. These
sites, which are dated to 2 to 1.5 mya, measure only 10 to 20 m (33 to 66 ft) in diam-
eter, but they are littered with fossilized animal bones (Figure 11.10). The densities of
animal bones in the archeological sites are hundreds of times higher than those in the
surrounding areas or in modern savannas. The bones belong to a wide range of ani-
mal species, including bovids (such as present-day antelope and wildebeests), pigs,
equids (horses), elephants, hippopotamuses, rhinoceroses, and a variety of carnivores
(Figure 11.11).
Along with these bones, Mary Leakey found several kinds of stone artifacts: cores,
flakes, battered rocks that may have been used as hammers or anvils, and some stones
that show no signs of human modification or use. The artifacts were manufactured
In this site
FIGURE 11.10
map for one
level of Bed I at Olduvai Gorge, most
0 1 2 3 4 ft
of the bones are from an elephant, and
0 1m
tools are shown in black.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 295
Elephant
Giraffe
Hippopotamus
Rhinoceros
Primate
Carnivore
Equid
Suid
Bovid
0 5 10 15 20 25 30 35 40 45 50
Number of individuals (%)
The bones of many different mammals were found at one archaeological site
FIGURE 11.11
at Olduvai. Bovids (which include antelope, gazelles, sheep, goats, and cat-
tle) clearly outnumber all other taxa.
from rocks that came from a number of different spots in the local area; some were
made from rocks found several kilometers away (Figure 11.12).
The association of hominin tools and animal bones does not necessarily mean
that early hominins were responsible for these bone accumulations.
It is easy to jump to the conclusion that the association of hominin tool and ani-
mal bones means that the Oldowan toolmakers hunted and processed the prey whose
bones we find at these sites. If that is our assumption, these may have been sites where
Oldowan hominins lived, like modern foragers’ camps, or they may have been butch-
ery sites where Oldowan hominins processed carcasses but did not live. However, there
are also other possibilities. Bones may have accumulated at these sites without any help
from early hominins. The bones may have been deposited there by moving water
(which has now disappeared) or by other carnivores, such as hyenas. These sites also
might be where many animals died of natural causes. Hominins might have visited the
sites after the bones accumulated, perhaps hundreds of years later, and left their tools
Naibor
Soit
This map
FIGURE 11.12
of Olduvai
5 km Gorge shows the major Bed I sites.
Early hominid Some of the tools were made of
sites Volcanic quartzite from Naibor Soit, others
highlands were made from gneiss from Kelogi,
Kelogi
and some used lava cobbles that came
from streams in the volcanic highlands
to the south.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 296
behind. Thus, the association of stones and bones is not enough to prove that
Oldowan hominins hunted or ate meat.
The study of taphonomy provides one means of resolving questions about what
happened at these sites. Taphonomy is the study of the processes that produce archae-
ological sites. Taphonomists examine the characteristics of contemporary kill sites—
spots where animals have been killed, processed, and eaten by various predators,
including contemporary human hunters. They monitor how each type of predator con-
sumes its prey, noting whether limb bones are cracked open for marrow, which bones
are carried away from the site, how human hunters use tools to process carcasses, and
how bones are distributed by predators at the kill site. These data enable archaeolo-
gists to develop a profile of the characteristics of kill sites created by different types of
predators. Taphonomists can also assess many of the same characteristics in archae-
ological sites. By comparing the features of archaeological and contemporary sites,
they are sometimes able to determine what happened at an archaeological site in the
past.
Taphonomic analyses at Olduvai Gorge suggest that the bones at most of these
sites were not accumulated by natural processes.
Taphonomic studies of the Olduvai sites tell us, first of all, that the bones were not
deposited by moving water. Animals sometimes drown as they try to cross a swollen
river or when they are swept away in flash floods (Figure 11.13). The bodies are car-
ried downstream and accumulate in a sinkhole or on a sandbar. As the bodies decom-
pose, the bones are exposed to the surrounding elements. The study of modern sites
shows that sediments deposited by rapidly moving water have a number of distinctive
characteristics. For example, such sediments tend to be graded by size because particles
of different size and weight sink at different spots. Sediments surrounding Olduvai sites
do not show any of the features characteristic of
sediments deposited by rapidly moving water.
Taphonomic analyses also tell us that the dense
concentrations of bones were not due to the deaths
of a large number of animals at one spot. Some-
times many animals die in the same place. In severe
droughts, for example, large numbers of animals
may die near water holes. Mass deaths usually
involve members of a single species, and there is
typically little mixing of bones from different car-
casses. By contrast, the bones at the Olduvai sites
come from a number of different species, and bones
from different carcasses are jumbled together.
At some sites, however, bone accumulations
do seem to be the product of natural processes, not
hominin activity. There is one site where the pat-
tern of bone accumulation is very similar to the
Sometimes large numbers of wildebeest drown when pattern of bone accumulations near modern hyena
FIGURE 11.13
trying to cross swollen rivers. dens. Hyenas often carry and drag carcasses from
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 297
kill sites to their dens so that they can feed their young and avoid competition with
other carnivores. At this site, there is no evidence of hominin activity, such as stone
tools or tool marks on bones.
The bones at these sites provide direct evidence of hominin activity. When carni-
vores gnaw on meat, their teeth leave distinctive marks on the bones. Similarly, when
humans use stone tools to butcher prey, their tools leave characteristic marks. Flaked-
stone tools have microscopic serrations on the edges and make very fine parallel
grooves when they are used to scrape meat away from bones (Figure 11.14). Many of
the animal bones at Olduvai show signs of stone tool use.
Bone
(a)
(b) (c)
The marks made on bone by teeth differ from the marks made by stone
FIGURE 11.14
tools. (a) The smooth surfaces of teeth leave broad, smooth grooves on
bones; the edges of stone tools, on the other hand, have many tiny, sharp points that leave fine parallel
grooves. Cut marks made by (b) carnivore teeth and (c) stone tools can be distinguished when they are
examined with a scanning electron microscope. These are scanning electron micrographs of 1.8-million-
year-old fossil bones from Olduvai Gorge.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 298
Hunters or Scavengers?
There has been controversy about whether the Oldowan hominins were hunters
or scavengers.
The archaeological evidence indicates that Oldowan hominins processed the carcasses
of large animals, and we assume that they ate the meat they cut from the bones. But
eating meat does not necessarily imply hunting. Some carnivores acquire meat by hunt-
ing, but many carnivores rely at least partly on scavenging. Scavengers steal kills from
other predators or rely on opportunistic discoveries of carcasses.
There has been considerable dispute about how early hominins acquired the
meat they ate. Some have argued that Oldowan hominins killed the prey found at
the archaeological sites; others have argued that Oldowan hominins could not have
captured large mammals, because they were too small, too poorly armed, and too
poorly encephalized. They contend that the Oldowan hominins were scavengers who
occasionally appropriated kills from other predators or collected carcasses they
found.
To resolve the debate about whether early hominins were hunters or scavengers,
we must first rethink popular conceptions of scavengers. Although scavengers have
an unsavory reputation, scavenging is not an occupation for the cowardly or the
lazy. Scavengers must be brave enough to snatch kills from the jaws of hungry com-
petitors, shrewd enough to hang back in the shadows until the kill is momentarily
left unguarded, or patient enough to follow herds and take advantage of natural
mortality. Studies of contemporary carnivores show that the great majority of scav-
enged meat is acquired by taking a kill away from another predator. Most preda-
tors respond aggressively to competition from scavengers. For example, lions
jealously guard their prey from persistent scavengers that try to steal bits of meat
or to drag away parts of the carcass. These contests can be quite dangerous (Fig-
ure 11.15).
(a) (b)
We also tend to think that some carnivores, like lions and leopards, only hunt, and
that others, like hyenas and jackals, only scavenge. But the simple dichotomy between
scavengers and hunters collapses when we review the data on the behavior of the five
largest African mammalian carnivores: lion, hyena, cheetah, leopard, and wild dog.
The fractions of meat obtained by scavenging vary from none for the cheetah, to 33%
for hyenas, with the others ranging somewhere in between. Contrary to the usual
stereotypes, the noble lion is not above taking prey from smaller competitors, includ-
ing female members of his own pride, and hyenas are accom-
plished hunters (Figure 11.16). For most large carnivores in
eastern Africa, hunting and scavenging are complementary
activities.
No mammalian carnivores subsist entirely by scaveng-
ing. It would be difficult for any large mammal to do so. For
one thing, many prey species create movable feasts, migrat-
ing in large herds over long distances. Although natural mor-
tality in these herds might make scavenging feasible, their
migratory habits eliminate this option. Mammalian carni-
vores cannot follow these migrating herds very far, because
they have dependent young that cannot travel long distances
(Figure 11.17). Only avian scavengers that can soar over
great distances, like the griffon vulture, rely entirely on scav- Male lions sometimes take kills from
enging (Figure 11.18). When migratory herds are absent, FIGURE 11.16
smaller carnivores and from female lions.
mammalian carnivores rely on other prey, such as waterbuck
and impalas. Natural mortality among resident species is not high enough to satisfy
the caloric demands of carnivores, so they must hunt and kill much of their own prey,
though they still scavenge when the opportunity arises.
Scavenging might be more practical if carnivores switched from big game to other
forms of food when migratory herds were not present. This is a plausible option for
early hominins. Most groups of contemporary foraging people rely heavily on gath-
ered foods—including tubers, seeds, fruit, eggs, and various invertebrates—in addi-
tion to meat. A few, such as the Hadza, obtain meat from scavenging as well as from
As dusk falls, three cheetah cubs wait for Vultures, which soar on thermals and have
FIGURE 11.17 FIGURE 11.18
their mother to return from hunting. enormous ranges, are the only carnivores
that rely entirely on scavenging.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 300
hunting. It is possible that early hominins relied mainly on gathered foods and scav-
enged meat opportunistically.
As we saw earlier, predators often face stiff competition for their kills. An ani-
mal that tries to defend its kill risks losing it to scavengers. For this reason, leopards
drag their kills into trees and eat the meat in safety. Other
predators, like hyenas, sometimes rip off the meaty parts of
the carcass, such as the hindquarters, and drag their booty
away to eat in peace. This means that limb bones usually
disappear from a kill site first, and less meaty bones, such
as the vertebrae and skull, disappear later or remain at the
kill site (Figure 11.19). If hominins obtained most of their
meat from scavenging, we would expect to find cut marks
made by tools mainly on bones typically left at kill sites by
predators, such as vertebrae. If hominins obtained most of
their meat from their own kills, we would expect to find
tool cut marks mainly on large bones, like limb bones. At
Olduvai Gorge, cut marks appear on all kinds of bones—
those usually left to scavengers and those normally monop-
olized by hunters (Figure 11.20). Moreover, some bones
After other predators have left, vultures
FIGURE 11.19 show cut marks on top of carnivore tooth marks, and other
consume what remains at the kill site.
bones show tooth marks on top of cut marks. Thus,
humans may sometimes have stolen kills from carnivores,
and vice versa. Taken together, the evidence suggests that early hominins acquired
meat by a combination of hunting and scavenging.
The numbers of cut marks on different kinds of bones vary at one site in the
FIGURE 11.20
Olduvai Gorge. There are more than twice as many pieces of bones from
large mammals (more than about 115 kg, or 250 lb) as there are from small mammals (less than
115 kg). The numbers in the figure are the percentages of all bones of a particular type that bear cut
marks. For example, 14% of all rib bones of larger mammals showed cut marks. In general, cut marks
are concentrated on bones that have the most meat.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 301
Some archaeologists, particularly the late Glyn Isaac, have suggested that the
dense accumulations of animal bones and stone tools found at some sites mark
the location of hominin home bases. They have speculated that early hominins
acquired meat by hunting or scavenging and then brought pieces of the carcass
home, where it could be shared. The dense collections of bones and artifacts were
thought to be the result of prolonged occupation of the home base (and sloppy
housekeeping). At one Olduvai Bed I site, there is even a circle of stones (Figure
11.22), dated to 1.9 mya, that is similar to the circles of stones anchoring the walls
of simple huts constructed by some foraging peoples in dry environments today.
Many Oldowan tools and bone fragments from a variety of prey species are found
at the same site.
However, a number of other observations are inconsistent with the idea that these
sites were home bases:
• Both hominins and nonhominin carnivores were active at many of the Olduvai
sites. Many of the bones at the Olduvai sites were gnawed by nonhominin carni-
vores. Sometimes the same bones show both tooth marks and cut marks, but some
show only the marks of nonhominin carnivores.
• Hominins and nonhominin carnivores apparently competed over kills. The bones A circle of
of nonhominin carnivores occur more often than would be expected on the basis FIGURE 11.22
stones at one
of their occurrence in other fossil assemblages or modern carnivore densities. Per- Olduvai Bed I site has been interpreted
haps the carnivores were killed (and eaten) when attempting to scavenge hominin by some archaeologists as the remains
kills or when hominins attempted to scavenge their kills. Hominins may not have of a simple shelter.
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 302
30
cially common when a small predator, like a cheetah,
25 makes a kill. The kill attracts many other animals, most
20 of which are able to displace the cheetah.
• The bones accumulated at the Olduvai sites are weath-
15
ered. When bones lie on the surface of the ground, they
10 crack and peel in various ways. The longer they remain
5 exposed on the surface, the greater the extent of weath-
ering. Taphonomists can calibrate the weathering
0
Fresh Heavily weathered process and use these data to determine how long fossil
Amount of weathering bones were on the ground before being buried. Some of
the bones at the Olduvai sites were exposed to the ele-
Many of the bones at Bed I sites at Oldu-
FIGURE 11.23 ments for at least four to six years (Figure 11.23).
vai are heavily weathered, suggesting that
they were deposited and exposed to the elements over a fairly long • The Olduvai sites do not show evidence of intensive
period of time. bone processing. The bones at these sites show cut
marks and tooth marks, and many bones were appar-
ently smashed with stone hammers to remove marrow.
However, the bones were not processed intensively, as
they are by modern foragers.
These observations are difficult to reconcile with the idea that the Olduvai sites
were home bases—places where people eat, sleep, tell stories, and care for their chil-
dren. First, foragers today do everything they can to prevent carnivores from entering
their camps. They often pile up thorny branches to fence their camps and keep dogs
that are meant to chase predators away. It is hard to imagine that early hominins could
have occupied these sites if lions, hyenas, and saber-toothed cats were regular visitors.
Second, bones at the Olduvai sites appear to have accumulated over a period of years.
Contemporary foragers usually abandon their home bases permanently after a few
months because the accumulating garbage attracts insects and other vermin. Even
though they revisit the same areas regularly, they don’t often reoccupy their old sites.
Finally, the fossilized bones found at Olduvai were not processed as thoroughly as
modern foragers process their kills.
Hominins may have brought carcasses to these sites and processed the carcasses
with flakes made from previously cached stones.
rocks for toolmaking at the sites of their kills. And they couldn’t leave their kills
unguarded while they went off to fetch their tools, lest a hungry scavenger steal their
supper. So they must have had to carry the meat to where their tools were kept or to
keep their tools with them all the time. Remember that these tools were fairly heavy,
and early hominins had no pockets or backpacks. Potts suggests that the best strategy
would have been to cache tools at certain places, and then to carry carcasses that
they had acquired to the nearest cache.
FURTHER READING
Conroy, G. 2005. Reconstructing Human Origins: A Modern Synthesis. 2nd ed. New
York: Norton.
Foley, R. 1987. Another Unique Species: Patterns in Human Evolutionary Ecology.
Essex, Eng.: Longman Scientific & Technical.
Kaplan, H., K. Hill, J. Lancaster, and A. M. Hurtado. 2000. A theory of human life
history evolution: Diet, intelligence, and longevity. Evolutionary Anthropology
9:156–185.
McGrew, W. C. 1992. Chimpanzee Material Culture: Implications for Human Evo-
lution. New York: Cambridge University Press.
Potts, R. 1984. Home bases and early hominins. American Scientist 72:338–347.
Schick, K. D., and N. Toth. 1993. Making Silent Stones Speak: Human Evolution and
the Dawn of Technology. New York: Simon & Schuster.
1. Who made the first stone tools? What kinds of evidence complicate this question?
2. What are the differences between collected, extracted, and hunted foods? How do
comparative data help us understand the unique foraging adaptations of modern
humans?
11_Oldowan Toolmakers [3p]:How Humans Evolved [5e] 10/6/08 3:32 PM Page 304
3. Why do complex foraging techniques favor slow development and long child-
hoods? What kinds of data enable us to draw inferences about developmental pat-
terns in early hominins?
4. Suppose we discovered that Oldowan foraging technology was very easy to mas-
ter and required little skill. How would that change your ideas about the selective
pressures acting on hominins? How would that change your ideas about who
made Oldowan tools?
5. Why do we associate food sharing with meat eating, rather than with vegetari-
anism?
6. What were Oldowan tools like, and what were they used for?
7. Researchers argue about whether Oldowan toolmakers were hunters or scav-
engers. What are the main arguments on each side of this debate?
8. How would it change our views of human evolution if we found convincing evi-
dence that the Oldowan hominins scavenged but did not hunt?
9. Some researchers have argued that Olduvai Bed I sites are the remains of hominin
home bases; others think that they are workplaces where hominins processed car-
casses. If you were an archaeologist, how would you go about testing this idea?
Think about the kinds of data you would need to collect to examine the merits of
each hypothesis.
10. We have discussed the Oldowan toolmakers at length, even though we don’t really
know who they were. Why is this a profitable exercise?
KEY TERMS
Chapter 12
Hominins of the Lower Pleistocene:
Homo ergaster
Morphology
Tools and Subsistence
Dispersal out of Africa
Eastern Asia: Homo erectus
Hominins of the Early Middle Pleistocene
(900 to 300 kya)
Africa and Western Eurasia:
Homo heidelbergensis
Hominins of the Later Pleistocene
(300 to 50 kya)
Eastern Eurasia: Homo erectus and
Homo heidelbergensis
Western Eurasia: The Neanderthals
Africa: The Road to Homo sapiens?
The Sources of Change
The Muddle in the Middle
A
bout 1.8 mya, a new kind of hominin appeared in Africa. These creatures,
who we will call Homo ergaster, were much more like modern humans than
the apelike hominins who preceded them. Some of them had large, robust
bodies with relatively long legs and short arms. They were fully committed to life on
the ground and may have been long-distance runners. Homo ergaster invented a new
kind of tool technology and probably learned to master fire and to hunt large game.
Although we can see much of ourselves in these creatures, some important differences
remain. They had smaller brains than we have, and their subsistence technology seems
to have been much less flexible than ours.
305
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 306
The Pleistocene is divided into three parts: the Lower, Middle, and Upper Pleistocene.
The Lower Pleistocene began about 1.8 mya, a date that coincides with a sharp cool-
ing of the world’s climate. The beginning of the Middle Pleistocene is marked by
sharply increased fluctuations in temperature and the first appearance of immense con-
tinental glaciers that covered northern Europe 900 kya, and its end is defined by the
termination of the penultimate glacial period about 130 kya. The Upper Pleistocene
ended about 12 kya when a warm, interglacial phase of the world climate began; this
warm period has persisted into the present (Figure 12.1).
Warmer 2
Colder
6 5 4 3 2 1 0
Millions of years ago
The pattern of world temperature over the last 6 million years. These esti-
FIGURE 12.1
mates are based on the ratio of oxygen-16 to oxygen-18 in cores extracted
from deep-sea sediments. During the Pliocene, world temperatures declined, and in the Pleistocene
climate fluctuations increased, especially during the Middle and Upper Pleistocene.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 307
Supraorbital
torus
Supraorbital
torus
Occipital
torus Zygomatic
Zygomatic arch
arch
0 5 cm
Homo ergaster skulls, like the skull of KNM-ER 3733 illustrated here, show a
FIGURE 12.2
mix of primitive and derived features. (Figure courtesy of Richard Klein.)
Homo ergaster appears in the African fossil record about 1.8 mya and
disappears about 0.6 mya.
Fossils of Homo ergaster have been found at several sites in Kenya (Lake Turkana,
Olorgesailie, and Ileret), as well as Konso-Gardula (Ethiopia), Daka (Ethiopia), Oldu-
vai Gorge (Tanzania), and Swartkrans (South Africa). Figure 12.2 shows a very well-
preserved skull (labeled KNM-ER 3733) from Lake Turkana that was found in 1976
by a team led by Richard Leakey. When H. ergaster fossils first appear on the scene,
they are associated with Oldowan tools. Recent finds in Ileret, Kenya, suggest that H.
habilis and H. ergaster may have coexisted for almost half a million years in the
Turkana basin of Kenya.
Until recently, most paleontologists assigned these East African fossils to the
species Homo erectus on the basis of their similarities to fossils that had been discov-
ered in Indonesia at the end of the nineteenth century. However, a growing number of
paleontologists have come to believe that the African specimens are distinctive enough
to be assigned to a different species. According to the rules of zoological nomencla-
ture, the Indonesian specimens retain the original name H. erectus because they were
described first, and the African specimens must be given a new name. Advocates of this
view have given the African fossils the name Homo ergaster, or “work man.” We will
use H. ergaster to refer to African specimens of this group and H. erectus for the Asian
fossils. Keep in mind that it is not entirely clear how many species these fossils
represent.
Morphology
Skulls of Homo ergaster differ from those of both earlier hominins and modern
humans.
Skulls of Homo ergaster retain many of the characteristics of earlier hominins (Figure
12.3), including a marked narrowing behind the eyes, a receding forehead, and no
chin. H. ergaster also shows many derived features. Some of these are shared by mod-
ern humans, including a shorter and less prognathic face, a taller skull, and smaller
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 308
Sagittal crest
Zygomatic arch
Ramus
(a) (b)
Occipital
Skulls of torus
FIGURE 12.3
(a) Aus- Zygomatic Ramus
tralopithecus africanus, (b) Paran- arch
thropus robustus, (c) Homo
ergaster, and (d) modern Homo
Chin
sapiens. (Figures courtesy of
Richard Klein.) (c) (d)
jaws and postcanine teeth, and reduction in the number of roots on the upper pre-
molars. However, H. ergaster also has some derived features that we don’t see in ear-
lier hominins or modern humans. For example, H. ergaster has a horizontal ridge at
the back of the skull (occipital torus), which gives it a pointed appearance when viewed
from the side. It also has quite large browridges.
Many of the derived features of the skull of Homo ergaster may be related to diet.
These hominins were probably better adapted for tearing and biting with their
canines and incisors, and less suited to heavy chewing with their molars. Thus, all their
teeth are smaller than the teeth of the australopiths and paranthropines, but their
molars are reduced relatively more in comparison with their incisors. The large brow-
ridges and the point at the back of the skull may have been needed to buttress the skull
against novel stresses created by an increased emphasis on tearing and biting.
Homo ergaster may also have been better adapted for life in a drier environment.
In apes and earlier hominin species, the nose is relatively flat and the nostrils face for-
ward. Like modern humans, H. ergaster’s nostrils face downward, and it probably had
a projecting nose. This may have been an adaptation for preserving moisture during
periods of physical exertion. The nose is generally cooler than the body, and moisure
condenses inside the nose when we exhale. Richard Klein of Stanford University spec-
ulates that H. ergaster may also have been the first hominin species to have hairless
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 309
The postcranial skeleton of Homo ergaster is much more similar to the skeleton
of modern humans than to that of earlier hominins, but it still differs from ours
in interesting ways.
Fossils attributed to Homo ergaster differ considerably in size. The smallest ones,
like KNM-ER 42700 from Ileret, may have been the same size as Homo habilis and
the australopiths, while others were considerably taller. The most complete informa-
tion about the postcranial morphology of H. ergaster comes from a spectacular spec-
imen discovered by Kimoya Kimeu (Figure 12.4), the leader of the Koobi Fora
paleontological team. The find was made on the west side of Lake Turkana, the same
region where fossils of Australopithecus anamensis, Paranthropus aethiopicus, and
Kenyanthropus platyops were found. The skeleton, which is formally known as KNM-
WT 15000, belonged to a boy who was about 12 years old when he died. The skele-
ton provides us with a remarkably complete picture of the H. ergaster body; even the
delicate ribs and vertebrae are preserved (Figure 12.5). An extensive analysis of this
skeleton, coordinated by Alan Walker, tells us a lot about this youngster’s body and
his way of life.
Remember that earlier hominins were bipeds but still had long arms, short legs,
and other features suggesting that they spent a considerable amount of time in trees.
In contrast, KNM-WT 15000 had the same body proportions as people who live in
tropical savannas today: long legs, narrow hips, narrow shoulders, and barrel-shaped
chest. KNM-WT 15000 also had short arms, compared with earlier hominins. Taken
together, these features suggest that H. ergaster was fully committed to terrestrial life.
This skeleton and less complete bits of the skeletons of other H. ergaster individ-
uals tell us several other interesting things about them:
The fossil of
• Some were quite tall. KNM-WT 15000 stood about 1.625 m (5 ft 4 in.) in height. FIGURE 12.5
the Homo
If Homo ergaster growth patterns were comparable to those of modern humans, ergaster boy, KNM-WT 15000, found
this boy would have been about 1.9 m (6 ft) tall when he was fully grown. How- on the west side of Lake Turkana, is
ever, he was also robust and heavily muscled. Think of him as a young shooting amazingly complete. (Photograph cour-
guard or a small forward. tesy of Alan Walker.)
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 310
• Sexual dimorphism was reduced. Homo ergaster males were about 20% to 30%
larger than females, making H. ergaster much less dimorphic than the apelike
hominins, but still more dimorphic than modern humans.
• They may not have had spoken language. The vertebral canal in the thoracic (mid-
dle) region of the back is much larger in modern humans than it is in apes, and it
contains a proportionally thicker spinal cord. Detailed studies of the neu-
roanatomy in the thoracic region suggest that all of the extra nerves that enlarge
the spinal cord innervate the muscles of the rib cage and diaphragm. The thoracic
vertebrae of KNM-WT 15000 are comparable to those of other primates. If this
young male is typical, then H. ergaster would have had less precise control over
the muscles of their rib cage and diaphragm than modern humans do. Anatomist
Ann MacLarnon of Roehampton University argues that increased motor control
of the diaphragm and thoracic muscles is an adaptation that allows people to reg-
ulate complex timing of breathing associated with speech.
• Homo ergaster was fully committed to life on the ground and was the first
hominin that could run for long distances. Compared to most other mammals,
modern humans are not good sprinters. However, we are able to outrun all but a
few species over distances of several kilometers. Biologist Dennis Bramble of the
University of Utah and anthropologist Daniel Lieberman of Harvard University
have recently argued that features of the H. ergaster boy like long legs, narrow
hips, and a barrel-shaped chest are evidence that the capacity for long-distance
running first appeared in this species. They believe that this talent may have been
useful in long-distance scavenging and hunting in open country.
Homo ergaster may have developed more slowly than early hominins, but more
rapidly than modern humans.
Remember from Chapter 10 that Christopher Dean and his colleagues used esti-
mates of enamel growth rate to show that australopiths developed relatively quickly.
Using the same methods, they estimate that H. ergaster developed more slowly than
australopiths but still faster than modern humans. If this interpretation is correct, then
H. ergaster did not have as long a childhood as modern humans have, suggesting that
learning did not play as important a role in the lives of these creatures.
Homo ergaster made fancier tools than earlier hominins had made.
The earliest fossils of Homo ergaster both in Africa and in Eurasia are associated with
Oldowan tools, the same tools that we described in Chapter 11. However, sometime
5 10 cm
0 between 1.6 and 1.4 mya in Africa, H. ergaster added a new and more sophisticated
4 in.
tool to its kit. This totally new kind of stone tool is called a biface. To make a biface,
Acheulean
FIGURE 12.6 the toolmaker strikes a large piece of rock from a boulder to make a core and then
hand axes, like
the two shown here, were teardrop- flakes this core on all sides to create a flattened form with a sharp edge along its entire
shaped tools created by removing flakes circumference. The most common type of biface, called a hand ax, is shaped like a
from a core. The smallest ones would fit teardrop and has a sharp point at the narrow end (Figure 12.6). A cleaver is a lozenge-
in the palm of your hand, and the largest shaped biface with a flat, sharp edge on one end; a pick is a thicker, more triangular
ones are more than 0.6 m (2 ft) long. biface. Bifaces are larger than Oldowan tools, averaging about 15 cm (6 in.) in length
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 311
and sometimes reaching 30 cm (12 in.). Bifaces are categorized as Mode 2 technology.
Paleoanthropologists call the Mode 2 industries of Africa and western Eurasia (that
is, Europe and the Middle East) the Acheulean industry after the French town of Saint-
Acheul, where hand axes were first discovered. The oldest Acheulean tools found at
West Turkana date to about 1.6 mya, just after H. ergaster first appeared in Africa. It
is important to understand that Oldowan tools do not disappear when Acheulean tools
make their debut. Homo ergaster continued making simple Mode 1 tools, perhaps
when they needed a serviceable tool in a hurry—as foraging people do today.
The standardized form of hand axes and other Mode 2 tools in the Acheulean
industry suggests that toolmakers had a specific design in mind when they made each
tool. The Mode 1 tools of the Oldowan industry have a haphazard appearance; no
two are alike. This lack of standardization suggests that makers of Oldowan tools sim-
ply picked up a core and struck off flakes; they didn’t try to create a tool with a par-
ticular shape that they had in mind beforehand. They may have done this because the
flakes were the actual tools.
It is easy to see how a biface might have evolved from an Oldowan chopper by
extending the flaking around the periphery of the tool. However, Acheulean tools are
not just Oldowan tools with longer edges; they are designed according to a uniform
plan. Hand axes have regular proportions: the ratio of height to width to thickness is
remarkably constant from one ax to another. Homo ergaster must have started with
an irregularly shaped piece of rock and whittled it down by striking flakes from both
sides until it had the desired shape. Clearly, all the hand axes would not have come
out the same if their makers hadn’t shared an idea for the design.
If hand axes were designed, what were they designed for? The answer to this ques-
tion is not obvious, because hand axes are not much like the tools made by later peo-
ples. A number of ideas about what hand axes were used for have been proposed:
1. Butchering large animals. Homo ergaster acquired the carcasses of animals like
zebra or buffalo either by hunting or by scavenging, and then used a hand ax as
a modern butcher would use a cleaver—to dismember the carcass and cut it into
useful pieces.
2. Digging up tubers, burrowing animals, water. Contemporary foragers in savanna
environments spend a lot of their time digging up edible tubers. Although mod-
ern peoples generally use sharpened sticks for digging, some archaeologists have
suggested that H. ergaster may have used hand axes for this purpose. Digging
tools would also have been useful for capturing burrowing animals like warthogs
and porcupines, and for making wells to acquire water.
3. Stripping bark from trees to get at the nutritious cambium layer underneath.
4. Hurling at prey animals.
5. Dispensing flake tools. Hand axes weren’t tools at all. Instead they were “flake
dispensers” from which H. ergaster struck flakes to be used for many everyday
purposes.
Although we are not certain how hand axes were used, two kinds of evidence sup-
port the hypothesis that they were heavy-duty butchery tools. Kathy Schick and
Nicholas Toth, whose investigations into the function of Oldowan tools we discussed
in Chapter 11, have also done experiments using Acheulean hand axes for each of the
tasks just listed. From these experiments, Schick and Toth conclude that hand axes are
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 312
best suited to butchery. The sharp end of the hand ax easily cuts through meat and sep-
arates joints; the rounded end provides a secure handle. The large size is useful because
it provides a long cutting edge, as well as the cutting weight necessary to be effective
for a tool without a long handle. Schick and Toth’s results are supported by the work
of University of Illinois at Chicago paleoanthropologist Lawrence Keeley, who per-
formed microscopic analysis of wear patterns on a small number of hand axes. He con-
cluded that the pattern of wear is consistent with animal butchery.
Evidence from Olorgesailie, a site in Kenya at which enormous numbers of hand
axes have been found, indicates that these axes may also have served as flake dis-
pensers. A team led by Richard Potts of the Smithsonian Institution discovered most
of the fossilized skeleton of an elephant, along with numerous small stone flakes dated
to about 1 mya. Chips on the edges of the flakes suggest that they were used to butcher
the elephant, and the elephant bones show cut marks made by stone tools. Careful
examination of the flakes reveals that they were struck from an already flaked core,
such as a hand ax, not from an unflaked cobble. Moreover, when the flakes were
removed from the hand axes, they did not leave a sharper hand-ax edge. Taken
together, these data suggest that H. ergaster struck flakes from hand axes and used
both hand axes and flakes as butchery tools.
There is relatively little change over space and time in the Acheulean tool kit from
its first appearance about 1.6 mya until it was replaced around 300 kya. Amazingly
enough, Acheulean tools that were made half a million years apart are just as similar
as tools made at about the same time at sites located thousands of miles apart. Essen-
tially the same tools were made for more than 1 million years. In fact, as we will see
shortly, the Acheulean industry was longer-lived than H. ergaster itself, which disap-
peared from the fossil record in Africa about 1 mya. Most anthropologists assume that
the knowledge necessary to make a proper hand ax was passed from one generation
to the next by teaching and imitation. It is a remarkable notion that this form of
knowledge might have been faithfully transmitted and preserved for so long in a small
population of hominins spread from Africa to eastern Eurasia.
One line of evidence that these hominins ate meat comes from the skeleton of a
Homo ergaster woman (KNM-ER 1808) that was discovered at Koobi Fora near Lake
Turkana in northern Kenya by Alan Walker and his colleagues (Figure 12.7). The long
bones of this woman, who died about 1.6 mya, are covered with a thick layer of abnor-
mal bone tissue. This kind of bone growth is symptomatic of vitamin A poisoning (Fig-
ure 12.8). How could a hunter-gatherer get enough vitamin A to poison herself? The
most likely way would be to eat the liver of a large predator, like a lion or a leopard.
Much of the
FIGURE 12.7 The same symptoms have been reported for Arctic explorers who ate the livers of polar
skeleton of a
Homo ergaster female (KNM-ER bears and seals. If this woman’s bones were deformed because she ate a large preda-
1808) was discovered at Koobi Fora. tor’s liver, then we can assume that H. ergaster ate meat. Of course, we don’t know
The bones marked in red are those that how this woman obtained the liver that poisoned her. She might have scavenged the
were recovered. (Diagram by Alan liver from a predator’s carcass, or she might have killed the predator in a contest over
Walker.) a kill.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 313
Homo ergaster apparently ate meat, but we do not know how they acquired it or
how the meat was processed. There is considerable controversy about whether H.
ergaster obtained meat from scavenging or from hunting. The issues and evidence
about hunting are similar to the controversy surrounding hunting by Oldowan
hominins. As we noted earlier, the bones of the elephant found at Olorgesailie show
signs of cut marks from stone tools. It is not clear how hominins obtained the elephant
carcass. The archaeological evidence indicates that the site was not repeatedly used by
hominins or other animals. Perhaps the hominins killed the elephant and butchered it
at the kill site. Perhaps they came upon the body of the elephant after it had died and
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 314
i. Oncospheres develop
into larvae in the muscle
3. Oncospheres hatch,
penetrate intestinal 5. Head of the
wall, and circulate tapeworm
to musculature attaches to
the intestine
butchered it there. It does not seem likely that they stole the kill from another preda-
tor, because there are no tooth marks on the bones.
There is also some dispute about when early hominids gained control over fire.
Archaeologists have found areas of baked earth at two Kenyan sites dated to about
1.5 mya, but it is not certain if this represents the remains of a campfire or a slow-
burning natural fire. Excavations at Swartkrans Cave in South Africa have yielded
thousands of fragmented fossil bones of antelope, zebras, warthogs, baboons, and
Paranthropus robustus. Burnt bones have been found at 20 different levels dated from
1.5 to 1 mya. Oldowan tools and H. ergaster fossils were also found on some of these
levels. To determine whether these bones had been burnt in campfires, C. K. Brain of
the Transvaal Museum in Pretoria and Andrew Sillen of the University of Cape Town
compared the fossils to modern antelope bones burned at a range of temperatures.
They found that the very high temperatures characteristic of long-burning campfires
produce clear changes in the microscopic structure of bone—changes that can also be
seen in the burnt fossils at Swartkrans.
Stronger evidence for the use of fire comes from the site of Gesher Benot Ya’aqov
in northern Israel which is dated to 790 kya. Carefully sifting through thousands of
of artifacts, researchers identified a number of burned seeds, wood, and flint artifacts.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 315
The flint artifacts were concentrated in particular areas, which may have been hearths.
The burned items represent a small fraction of the material at the site, which makes
it unlikely that it was the product of a natural wildfire that swept through the site.
It is even harder to determine whether hominins used fire to cook their food.
Richard Wrangham of Harvard University has pointed out that cooking makes meat
easier to chew and digest. Moreover, tubers must be cooked before they are eaten. The
use of fire to cook meat or tubers might have enhanced the foraging efficiency of H.
ergaster. This might have been an important adaptation for these creatures, who had
to support the higher energetic costs of their larger brains.
Homo ergaster occupied almost all of Africa and extended its range into
Eurasia.
Boxgrove
Trinchera Mauer Dmanisi
Dolina
Zoukoudian
Petral0na
Bose Basin
Bodo
The locations of fossil
FIGURE 12.10
Konso-Gardula and archaeological
Olduvai sites mentioned in the text. Homo ergaster is
Lake Turkana
and Ndutu
found in Africa and Europe from 1.8 mya until
Kabwe Solo River about 1 mya. Homo heidelbergensis is found at
sites in Africa and Europe that date to between
800 and 300 kya. Only Homo erectus is found
in east Asia during this period. The Bose Basin
Swartkrans
has yielded Mode 2 tools but no hominin fossils,
so the identity of the toolmakers is unknown.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 316
medieval town called Dmanisi (Figure 12.11) discovered a hominin lower jaw and
Oldowan tools. In 1999, when an international team of researchers, led by the late Leo
Gabunia of the Republic of Georgia National Academy of Sciences, uncovered two
nearly complete crania at Dmanisi that are very similar to African specimens of H.
ergaster (Figure 12.12). Later, the team discovered a very well-preserved cranium and
associated mandible of a subadult with a very small brain (700 cm3), which shows
striking similarities to the recently discovered specimen from Ileret, Kenya. In 2007,
David Lordkipanidze and his colleagues published the first detailed descriptions of
postcranial material from one adolescent and three adults. The Dmanisi fossils are
associated with more than 1,000 Oldowan (Mode 1) choppers, scrapers, chopping
tools, and flakes. The residents of Dmanisi manufactured their tools from local mate-
rials but clearly preferred fine-grained stones, such as quartzite and basalt, for mak-
ing tools. Radiometric dating of the basalt underlying the
fossils guarantee that they are less than 1.8 million years old.
The magnetic polarization of the surrounding sediments
indicates that they are older than 1.2 million years old. The
associated fauna suggest that the hominin fossils were
deposited during the early part of this interval.
Lordkipanidze and his colleagues emphasize that the
Dmanisi hominins display a mix of primitive and derived fea-
tures. They have relatively small brains (600–775 cm3). Cal-
culations based on the relationship between brain size and
body size indicate that the Dmanisi hominins fall at the lower
end of the distribution for African H. ergaster, and appear
more like H. habilis and the australopiths. They also retain
primitive features in certain elements of their shoulder mor-
phology. In modern humans, the elbow joint is rotated so
One of the crania found at Dmanisi, a site
FIGURE 12.12 that our palms face inward when our arms hang by our
in the foothills of the Caucasus Mountains
in the Republic of Georgia. The Dmanisi fossils date to 1.2 to 1.8 sides. In the Dmanisi hominins, the palms would be oriented
mya and are very similar to Homo ergaster fossils of the same age more toward the front of the body. The lower-limb mor-
from Africa. The Dmanisi site is at latitude 41° north, well out of the phology shows a number of derived features. The Dmanisi
tropics, indicating that Homo ergaster was able to adapt to a wider hominins have the same limb proportions as modern
range of habitats than previous hominins were. humans, and their lower limbs and feet were well suited for
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 317
long-distance walking and running. For example, they have relatively high
femoral/tibia and humeral/femoral ratios, well-developed arches in their feet, and an
adducted big toe (that is, it is in line with the rest of the toes).
Fossils quite similar to what we now call Homo ergaster were unearthed by Eugène
Dubois near the Solo River in Java during the nineteenth century (Figures 12.10 and
12.13). Dubois named the species Homo erectus, or “erect man.” Dubois’s fossils were
very poorly dated, and for many years they were believed to be about 500,000 years
Eugène
old. In the 1990s, however, Carl Swisher and Garniss Curtis, both from the Berkeley FIGURE 12.13
Dubois was a
Geochronology Center, used newly developed argon–argon dating techniques to
Dutch anatomist who discovered the
determine the age of small crystals of rock from the sites at which Dubois’s fossils had first Homo erectus fossil at a site called
originally been found. Their analyses indicate that the two H. erectus sites in Java were Trinil, in what is now Indonesia.
actually 1.6 to 1.8 million years old, a million years older than previous estimates had
suggested. However, because Dubois’s excavations were made 100 years before geol-
ogists returned to assess the stratigraphic provenance of the finds, we need to treat
these dates with some caution.
The main morphological differences between Homo erectus and Homo ergaster
lie in the skull. In H. erectus, the face is more massive, the walls of the cranium are
thicker, the cranium is lower and less domed, the browridges are more pronounced,
the sides of the skull slope more steeply, the occipital torus is more pronounced, and
there is a sagittal keel (a longitudinal V-shaped ridge along the top of the skull) (Fig-
ure 12.14). As you may recall from Chapter 11, the sagittal crest expands the area of
attachment of the temporalis muscles in paranthropines. The sagittal keel does not
serve this function in H. erectus. In fact, we don’t know what its function was.
Sagittal keel
Homo erectus was shorter and stockier than Homo ergaster. Femurs from north-
ern China indicate that adult males were only about 1.6 m tall (5 ft 3 in.), about a head
shorter than KNM-WT 15000 would have been.
Homo erectus is known from sites in Java and China, and persisted until about
30 kya. Over this period, H. erectus changed very little. This species does not show
the same increase in cranial capacity, technological sophistication, and behavioral flex-
ibility that characterizes its contemporary hominin populations in Africa and western
Eurasia. Even the most recent H. erectus fossils from eastern Asia are less like mod-
ern humans than are the African specimens of H. ergaster, which are 1 million years
older.
Nonetheless, the presence of Homo erectus fossils in eastern Asia suggests that this
species mastered difficult environmental challenges. Somewhere between 500 and 250
kya, H. erectus occupied a cave at Zoukoudian (also Zhoukoudian or Choukoutien)
near Beijing. Fossil-pollen samples indicate that H. erectus occupied the cave only dur-
ing the warmer interglacial periods. However, warm is a relative term; the climate of
Zoukoudian during the interglacial periods was probably something like that of
Chicago today. Imagine living through a winter in Chicago without a roof over your
head or clothes to keep you warm. To survive the rigors of cold temperate winters, H.
erectus may have relied on fire and taken shelter in caves like the one at Zoukoudian.
Even this, however, might have been a challenge. The cave at Zoukoudian is littered
with the bones of large game, particularly two species of deer. It is clear that they were
brought to the cave by a predator, perhaps H. erectus. However, there are also fos-
silized hyena bones and coprolites (fossilized feces) in these caves. Many of the ani-
mal bones show evidence of hyena tooth marks, and most of the faces of the hominins
are absent. Thus, it is possible that H. erectus was sometimes the prey at Zoukoudian
rather than the predator.
were covered with dense bamboo forests. Bamboo is unique among woods because it
can be used to make sharp, hard tools suitable for butchering game. H. erectus may
not have made hand axes because they didn’t need them. The recent discovery of a
large number of hand axes in the Bose Basin in southern China may support this view
(Figure 12.15). About 800 kya, a large meteor struck this area and set off fires that
destroyed a wide area of the forest. Grasslands replaced the forest. Although grass-
lands predominated, the residents of this area made Mode 2 tools like those seen in
Africa around the same time. Before the forest was destroyed and after it regenerated,
the inhabitants of this area made Oldowanlike Mode 1 tools. This evidence suggests
that hand axes were an adaptation to open-country life. However, there are no
hominin fossils associated with these tools, so we don’t know whether H. erectus actu-
ally made them. It is possible that the Bose hand axes were made by members of more
technologically advanced immigrant populations from farther west.
During the Middle and Upper Pleistocene there were many long, cold glacial periods
punctuated by short, warmer interglacial periods. Figure 12.16 shows estimates of
global temperatures for this time period. Notice that the world climate has fluctuated
wildly in the last 700,000 years. From geological evidence, we know that during
the cold periods, glaciers covered North America and Europe, and arctic conditions
prevailed. These cold periods were intermittently interrupted by shorter warm periods
during which the glaciers receded and the forests returned.
During glacial periods, the world was dry, and Africa and Eurasia were isolated
from each other by a massive desert. During interglacial periods, the world was
much wetter, and animals moved from Africa to Eurasia.
Ocean water
increase in
16O/18O
Years ago
0 Top of core
13,000
24,000
59,000 Upper
71,000 Pleistocene
128,000
186,000
245,000
303,000
339,000
Core depth (cm)
362,000
Middle
Pleistocene
423,000
478,000
524,000
565,000
620,000
659,000
689,000
726,000
1,200 736,000
most animal species. This corresponds to what we know from the fossil record: There
was little movement of animal species between Africa and Asia during glacial periods.
Instead, animal species moved mainly east and west across Eurasia. When the world
was warmer, grasslands and savannas replaced most deserts, and animals were able
to move between Africa and Eurasia much more easily. The fossil record indicates that
animal species moved mainly from Africa to Eurasia, not vice versa. As we will see
later, this fact has important implications for understanding human evolutionary
history.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 321
Boreal forest
Forest & woodland
Wooded savanna & scrub
Grassland
Semidesert
Extreme desert
Steppe & tundra
Cold desert
Ice
(a) A recon-
FIGURE 12.17
struction of
(b) Glacial conditions (~20,000 years BP) biological habitats about 7 kya during
the warmest, wettest part of the pres-
ent interglacial period. Much of Eurasia
and Africa were covered with forest and
were connected by a broad swath of
grassland across northern Africa and
southwestern Eurasia. (b) The habitats
during the coldest, driest part of the
last glacial period, about 20 kya. There
was very little forest cover. Grassland
and scrub predominated in central
Africa and Southeast Asia. Northern
Eurasia was covered with cold, dry
steppe and desert. Central and southern
Africa were separated from Eurasia by a
band of extreme desert across northern
Africa, the Arabian Peninsula, and cen-
tral Asia.
It was in this chaotic, rapidly changing world that natural selection reshaped the
hominin lineage once again. Homo ergaster was transformed into a smarter and more
versatile creature. We will see that hominins were present in Africa and Eurasia for
most of the Middle Pleistocene, and that hominins living in different parts of the world
were morphologically distinct. H. erectus lived in eastern Asia until perhaps 30 kya;
and a variety of larger-brained forms ranged through Africa and western Eurasia.
About 300 kya, hominins in Africa and western Eurasia began to develop more sophis-
ticated technology and behavior. This process of change continued, particularly in
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 322
Africa, until the behavior and technology of hominins became indistinguishable from
those of modern humans.
Although many of these events can be relatively well established in the fossil and
archaeological record, there is considerable controversy about the phylogeny of the
human lineage during this period. There are fierce debates about how many species of
hominins there were during this period, and about how different species or forms were
related. We will turn to these controversies at the end of the chapter. But first you need
to learn something about the hominins of the Middle Pleistocene.
Sometime during the first half of the Middle Pleistocene (900 to 130 kya),
hominins with larger brains and more modern skulls appeared.
Hominins with substantially larger brains and more modern skulls appeared in Africa
and western Eurasia during the first half of the Middle Pleistocene. Figure 12.18 shows
M
Higher, more
rounded
skull
Massive
browridges
Occipital K
torus
Kabwe
Higher, more
rounded
skull
Some-
FIGURE 12.18
time
Massive
between 800 and 500 kya,
browridges
hominins with higher, more rounded
crania and larger brains first appear
in the fossil record; and by 400 kya,
Homo heidelbergensis was common
in Africa and western Eurasia.
Shown here are two Homo heidel-
bergensis fossils, one from Kabwe
Occipital
(sometimes called Broken Hill) in
torus
Zambia, and the second from
Petralona in Greece. Both are
O
approximately 400,000 years old.
(Figure courtesy of Richard Klein.) Petralona
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 323
two nearly complete crania from this period: the Petralona cranium found in Greece and
the Kabwe (or Broken Hill) cranium from Zambia. These individuals had substantially
larger brains—between 1,200 and 1,300 cc—in relation to their body size than Homo
ergaster had. The skulls also share a number of derived features with modern humans,
including more vertical sides, higher foreheads, and a more rounded back. However,
they also retained many primitive features, such as a long, low skull; very thick cranial
bones; a large prognathic face; no chin; and very large browridges. Their bodies were
still much more robust than modern human bodies are. Fossils with similar character-
istics have been found at other sites in Africa (including Ndutu in Tanzania and Bodo
in Ethiopia) and western Eurasia (including Mauer in Germany and Boxgrove in Eng-
land) that date to the same period (see Figure 12.10). There has been no sign of these
kinds of fossils in eastern Asia during this period, although they do appear later.
Traditionally, paleoanthropologists have referred to these larger-brained, more
modern-looking hominins as archaic Homo sapiens. As we will explain shortly, how-
ever, nowadays only a minority of paleoanthropologists believe that these creatures
should be classified as members of H. sapiens. There is considerable controversy about
whether these hominins belonged to one species or several, and among the anthro-
pologists who believe that there were several species, there is considerable disagree-
ment over which specimens belong to which species. This dispute, which is far from
resolved, creates a quandary about what we should call this group of hominins. Here
we adopt what is perhaps the most common opinion and use the name Homo hei-
delbergensis to refer to all of the Middle Pleistocene hominins of Africa and western
Eurasia. (The name heidelbergensis is taken from a specimen that was found near the
German city of Heidelberg.) Keep in mind, though, that we are not at all sure that this
category actually represents a single biological species.
Scientists are uncertain about when or where Homo heidelbergensis first appeared.
One candidate for the oldest H. heidelbergensis specimen is the cranium recently found
at Buia in Eritrea. This skull has several derived features that are associated with H.
heilelbergensis, but a small, H. ergaster–sized braincase. A second possibility is the
Kabwe fossil found in Zambia. Although it was originally dated to about 130 kya,
more recent work has pushed the age of this fossil back to about 800 kya. A third can-
didate comes from Trinchera Dolina (also called Gran Dolina) in the Sierra de Ata-
puerca in northern Spain. This site has yielded a number of hominin fossils, including
part of an adolescent’s lower jaw and most of an adult’s face. Although the Trinchera
Dolina fossils are too fragmentary to provide an estimate of endocranial volume, they
exhibit a number of facial features that are seen in more modern hominins. These fos-
sils have been dated to about 800 kya by means of paleomagnetic methods, but rodent
fossils found at the same site indicate a more recent date, perhaps 500 kya. Similar
uncertainty afflicts our estimates of the ages of the other early H. heidelbergensis fos-
sils, so the best that we can do is to bracket the first appearance of these creatures
between 800 and 500 kya.
The tools used by early Homo heidelbergensis are similar to those used by Homo
ergaster. Tool kits are dominated by Acheulean hand axes and other core tools at most
sites, but in some cases the hand axes are more finely worked.
The first solid evidence for hunting big game comes from this period. On the island
of Jersey, off the coast of France, the remains of a large number of fossilized bones
from mammoths and woolly rhinoceroses have been found at the base of a cliff. Some
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 324
of the bones come from adults, which were too big to be vulnerable to most preda-
tors. The carcasses clearly have been butchered with stone tools. In some instances, the
skull cavity has been opened, presumably to extract the brain tissue. At some places
on this site, animal bones have been sorted by body parts (heads here, limbs there, and
so on). All this suggests that H. heidelbergensis drove the animals
Flake the margin of the core: over the headland, butchered the carcasses, and ate the meat.
More evidence for hunting comes from three wooden spears
that were found in an open-pit coal mine in Schoningen, Germany.
This site is dated to about 400 kya. Anthropologists are confident
that these were throwing spears because they closely resemble
modern javelins. They are about 2 m (6 ft) long and, like modern
throwing spears, are thickest and heaviest near the pointed end,
gradually tapering to the other end. Some modern people use sim-
ilar spears for hunting, and it seems plausible that H. heidelber-
gensis used them in the same way. This hypothesis is strengthened
(a) (b)
by the fact that anthropologists found the bones of hundreds of
Prepare the surface of the core: horses along with the spears, and many of the bones show signs of
having been processed with stone tools.
(c) (d) About 300 kya, and then slightly later in western Eurasia,
Remove Levallois flake:
hominins in Africa shifted to a new stone tool kit.
Core Flake Beginning about 300 kya, hand axes became much less common
and were replaced by tools that were manufactured by the pro-
duction of sizable flakes, which were then further shaped or
retouched. Unlike the small, irregular Oldowan flakes that had been
struck from cobble cores, these tools were made from large, sym-
metrical, regular flakes via complicated techniques. One method,
called the Levallois technique (after the Parisian suburb where such
(e) (f)
tools were first identified), involves three steps. First, the knapper
The process of making a Levallois prepares a core with one precisely shaped convex surface. Then, the
FIGURE 12.19
tool. (a) The knapper chooses an knapper makes a striking platform at one end of the core. Finally,
appropriate stone to use as a core. The side and top views of the knapper hits the striking platform, knocking off a flake, the
the unflaked core are shown. (b) Flakes are removed from the shape of which is determined by the original shape of the core (Fig-
periphery of the core. (c) Flakes are removed radially from ure 12.19). A skilled knapper can produce a variety of different
the surface of the core, with the flake scars on the periphery
kinds of tools by modifying the shape of the original core. Such pre-
being used as striking platforms. Each of the red arrows repre-
pared core tools are classified as Mode 3 technology.
sents one blow of the hammer stone. (d) The knapper contin-
ues to remove radial flakes until the entire surface of the core Microscopic analyses of the wear patterns on Mode 3 tools
has been flaked. (e) Finally, a blow is struck (red arrow) to made during this period suggest that some tools were hafted
free one large flake (outlined in red). This flake will be used (attached to a handle). Hafting is an extremely important innova-
as a tool. (f) At the end, the knapper is left with the remains tion because it greatly increases the efficiency with which humans
of the core (left) and the tool (right). (Figure courtesy of can apply force to stone tools. (Try using a hammer without a han-
Richard Klein.) dle.) Homo heidelbergensis probably hafted pointed flakes onto
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 325
Hominins with larger brains and more rounded skulls have been found at several sites
in China that date to approximately 200 kya. The most complete and most securely
dated fossil is from the Jinniushan site, Yingkou, in northern China (Figure 12.20).
This specimen (Figure 12.21), which consists of a cranium and associated postcranial
bones, is similar to early Homo heidelbergensis fossils from Africa and Europe. Like
the crania found at Kabwe and Petralona, it shares a larger braincase (about 1,300 cc)
and more rounded skull with modern humans, but it also has massive browridges and
other primitive features. Other H. heidelbergensis fossils have been found in Dali in
northern China and Maba in southern China, and they are probably somewhat
younger than the Yingkou specimen. These fossils are associated with Oldowan-type
tools. It is unclear whether these hominins were immigrants from the west or the result
of convergent evolution in eastern Asia.
Homo heidelbergensis may have coexisted with Homo erectus in east Asia dur-
ing this period. Fossils of H. erectus that are between 200,000 and 300,000 years old
have been found both at Zoukoudian (Skull V) and at Hexian (also called He Xian),
in southern China. The fossils of other animals found in association with H. erectus
Omo Kibish
Locations of
Laetoli FIGURE 12.20
later Pleis-
Ngandong and tocene fossil and archaeological sites
Sambungmachan mentioned in the text. Homo erectus
Florisbad and Homo heidelbergensis are found in
Border Cave east Asia, Neanderthals in Europe, and
both Homo heidelbergensis and Homo
Klasies River mouth
sapiens in Africa.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:12 PM Page 326
Flat, receding
skull Large,
relatively
Strong browridges well-rounded
braincase
Broad, flat
face
Occipital
torus
Hominins with the characteristics of Homo
FIGURE 12.21
heidelbergensis appeared in east Asia later Reduced cm
than in western Eurasia. The specimen illustrated here, which is from molar
Yingkou in northern China, is approximately 200,000 years old. Yingkou (Jinniushan)
fossils at two sites in Java (Ngandong and Sambungmachan) are consistent with an age
of 250,000 to 300,000 years. However, the teeth of bovids (buffalo and antelope) asso-
ciated with the hominin fossils at these sites have been dated to about only 27 kya by
means of electron-spin-resonance techniques.
Borneo
New Guinea
Flores
Australia
dants of H. erectus, who became isolated on the island of Flores. They emphasize the
fact that the Flores hominins share a number of derived characters with H. erectus and
that a rich fossil record documents a long occupation of nearby Java by H. erectus. In
addition, simple Mode 1 tools like those associated with H. erectus have
been found on Flores and are dated to 800 kya. The small size of the Flo-
res hominins is thought to be an example of evolutionary dwarfism, which
occurs when animal populations are confined on islands with few predators
and limited supplies of food. Flores and the islands around it have always
been isolated from both Asia and Australia by wide stretches of ocean. As
a result, only a few species of large animals ever reached these islands, pro-
ducing a strange impoverished fauna that included Komodo dragons and
an even larger species of monitor lizard, a species of dwarf elephant, and,
of course, a dwarf hominin. Biologists think that natual selection favors
smaller body size on small islands because such islands typically have less
predation and more limited food supplies. If islands are isolated, there will
be little gene flow from continental populations, and the animals will adapt
to local conditions. On Flores, this process may have produced pint-sized
hominins and tiny elephants.
This interpretation is challenged by other experts who are convinced
Three-quarter view of
that the occupants of Ling Bua were members of a modern human popu- FIGURE 12.23
the skull of Homo flore-
lation with small stature. The tiny cranium from Ling Bua belonged to an siensis. The skull is very small but shares a number
individual who was afflicted with microcephaly, a disorder that is charac- of derived features with Homo erectus. The brain of
terized by reduced brain growth. Finally, some researchers emphasize the this creature was smaller than that of the smallest
unique combination of derived and ancestral traits in the Flores specimens australopiths.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 328
and suggest that they may be the descendants of a hominin that preceded Homo erec-
tus into Asia. The problem with this explanation is that we have no evidence of this
ancestral population.
A large sample of fossils from a recently excavated site in Spain provides evidence that
Homo heidelbergensis in Europe had begun to diverge from other hominin popula-
tions during the Middle Pleistocene. This site, which is called Sima de los Huesos (“Pit
of Bones”) is located in the Sierra de Atapuerca only a few kilometers from Trinchera
Dolina. There, paleoanthropologists excavated a small cave 13 m (43 ft) below the sur-
face and found 2,000 bones from at least 24 different individuals. These bones, which
date to about 300 kya, include several nearly complete crania, as well as many bones
from other parts of the body. Figure 12.24 shows one of the crania, labeled SH 5. Like
other fossils of H. heidelbergensis, these skulls mix derived features of modern
humans and primitive features associated with H. ergaster. However, the crania from
Sima de los Huesos also share a number of derived characters not seen in hominins
living at the same time in Africa. Their faces bulge out in the middle and have double-
arched browridges, and the backs of some of the skulls are rounded. The fossils from
Sima de los Huesos also have relatively large cranial capacities, averaging about
1,390 cc, close to the average value for modern humans. These characters are signifi-
cant because they are shared by Neanderthals, the hominins who dominate the Euro-
pean fossil record from 127 to 30 kya.
Not all of the Sima de los Huesos crania express all the Neanderthal features to
the same degree, and this variability has helped to settle an important question about
hominin evolution in Europe. Before the Sima de los Huesos fossils were discovered,
paleoanthropologists had been perplexed by the pattern of variation in Middle Pleis-
tocene hominins in Europe. Fossils from some European sites resemble Neanderthals;
fossils from other European sites that date to the same time period do not resemble
Neanderthals. It was not clear whether these differences reflected variability within a
The many
FIGURE 12.24
hominin fos-
sils found at Sima de los Huesos in
Spain provide evidence that hominins in
Europe began to evolve a distinctive cra-
nial morphology at least 300 kya. The
features—which include rounded brow-
ridges, a large pushed-out face, a skull
with a rounded back, and a large brain—
are important because they are shared
with the Neanderthals, the hominins
who dominate the European fossil
record during the Upper Pleistocene.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 329
single population of hominins or whether more than one type of hominin was pres-
ent in Europe during the Middle Pleistocene. The large and variable sample of indi-
viduals at Sima de los Huesos has resolved this issue. These individuals, who likely
belonged to a single population, vary in their expression of Neanderthal traits, and
this variation mirrors the kind of variation seen at different sites in Europe. This
evidence suggests that the features defining the Neanderthals had begun to evolve
in European populations by 300 kya, but that these traits evolved independently,
not as a single functional complex.
The Neanderthals appeared in western Eurasia about 130 kya. Increasing amounts of 18O
0
The last warm interglacial period lasted from about 130 kya to about 75
kya. For most of the time since then, the global climate has been colder—
100
sometimes much colder.
The data in Figure 12.25 give a detailed picture of global temperatures over
the last 123,000 years. These data are based on the 18O:16O ratios of different lay-
120
ers of cores taken from deep inside the Greenland ice cap. Because snow accumu-
lates at a higher rate than sediments on the ocean bottom do, ice cores provide
Colder Warmer
more detailed information on past climates than do ocean cores like those used to
construct the graph in Figure 12.1. You can see from Figure 12.25 that the end of Fluctuations in
FIGURE 12.25
the last warm interglacial period was marked by a relatively slow decline in tem- the ratios of 18O
16
to O over the last 123,000 years taken
perature beginning about 120 kya. During this interglacial period, the world gen-
from an ice core drilled in northern Green-
erally was substantially warmer than it is today. During this warm interglacial
land. These data indicate that, between
period, plants and animals were distributed quite differently than they are now.
120 and 80 kya, the world’s climate got
Plankton species currently living in subtropical waters (like those off the coast of colder and less stable. In ice cores like this
Florida) extended their range as far as the North Sea during the last interglacial one, a higher ratio of 18O to 16O indicates
period. Animals now restricted to the tropics had much wider ranges. Thus, the higher temperatures. (In deep-sea cores,
remains of a hippopotamus have been found under Trafalgar Square, which lies in more 18O relative to 16O indicates lower
the center of London. In Africa, rain forests extended far beyond their present temperatures.)
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 330
boundaries; and in temperate areas, broadleaf deciduous forests extended farther north
than they do today.
As the last glacial period began, sometime between 100 and 75 kya, the world
slowly cooled. In Europe, temperate forests shrank, and grasslands expanded. The gla-
ciers grew, and the world became colder and colder—not steadily, but with wide fluc-
tuations from cold to warm. When the glaciation was at its greatest extent (about 20
kya), huge continental glaciers covered most of Canada and much of northern
Europe. Sea levels dropped so low that the outlines of the continents were altered sub-
stantially: Asia and North America were connected by a land bridge that spanned the
Bering Sea; the islands of Indonesia joined Southeast Asia in a landmass called Sun-
daland; and Tasmania, New Guinea, and Australia formed a single continent called
Sahul. Eurasia south of the glaciers was a vast, frigid grassland, punctuated by dunes
of loess (fine dust produced by glaciers) and teeming with animals—woolly mam-
moths, woolly rhinoceroses, reindeer, aurochs (the giant wild oxen that are ancestral
to modern cattle), musk oxen, and horses.
• Large brains. The Neanderthal braincase is much larger than that of Homo
heidelbergensis, ranging from 1,245 to 1,740 cc, with an average size of about
1,520 cc. In fact, Neanderthals had larger brains than modern humans, whose
brains average about 1,400 cc. It is unclear why the brains of Neanderthals were
so large. Some anthropologists point out that the Neanderthals’ bodies were much
more robust and heavily muscled than those of modern humans, and they suggest
that the large brains of Neanderthals reflect the fact that larger animals usually
have bigger brains than smaller animals.
• More rounded crania than Homo erectus or Homo heidelbergensis (Figure
0 10 cm
12.26). The Neanderthal skull is long and low, much like the skulls of H. heidel-
bergensis, but relatively thin-walled. The back of the skull has a characteristic
The skulls
FIGURE 12.26 rounded bulge or bun and does not come to a point at the back like an H. erec-
of Nean-
derthals, like this one from Shanidar tus skull does. There are also detailed differences in the back of the cranium.
Cave, Iraq, are large and long, with large • Big faces. Like Homo erectus and Homo heidelbergensis, the skulls of Nean-
browridges and a massive face. (Figure derthals have large browridges, but they are larger and rounder and they stick out
courtesy of Richard Klein.) less to the sides. Moreover, the browridges of H. erectus are mainly solid bone,
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 331
while those of Neanderthals are lightened with many air spaces. The function of
Enamel Enamel
these massive browridges is not clear. The face, and particularly the nose, is enor-
mous: every Neanderthal was a Cyrano, or perhaps a Jimmy Durante.
Pulp Pulp
• Small back teeth and large, heavily worn front teeth. Neanderthal molars are
smaller than those of Homo ergaster. They had distinctive taurodont roots (Fig-
ure 12.27) in which the pulp cavity expanded so that the roots merged, partially
or completely, to form a single broad root. Neanderthal incisors are relatively large
and show very heavy wear. Careful study of these wear patterns indicates that
Neanderthals may have pulled meat or hides through their clenched front teeth. Nontaurodont Taurodont
tooth tooth
There are also microscopic, unidirectional scratches on the front of the incisors,
suggesting that these hominins held meat in their teeth while cutting it with a stone In Nean-
FIGURE 12.27
tool. Interestingly, the direction of the scratches suggests that most Neanderthals derthal
molars, the roots often fuse together
were right-handed, just as the Oldowan toolmakers were.
partially or completely to form a single
• Robust, heavily muscled bodies. Like Homo ergaster, Homo erectus, and Homo
massive taurodont root. The third root
heidelbergensis, Neanderthals were extremely robust, heavily muscled people (Fig-
is not shown.
ure 12.28). Their leg bones were much thicker than ours, the load-bearing joints
(knees and hips) were larger, the scapulae (shoulder blades; singular scapula) had
more extensive muscular attachments, and the rib cage was larger and more
barrel-shaped. All these skeletal features indicate that Neanderthals were very
sturdy and strong, weighing about 30% more than contemporary humans of the
same height. A comparison with data on Olympic athletes suggests that Nean-
derthals most closely resembled the hammer, javelin, and discus throwers and
shot-putters. They were a few inches shorter, on average, than modern Europeans
and had larger torsos and shorter arms and legs.
Neanderthal
Early modern
Early modern
vertical axis; crural index is plotted on the horizontal axis.
Eskimo
Mesolithic
(Europe)
Smaller values of the crural index are associated with shorter
(Israel)
limbs relative to body size. Populations in warm climates Lapp
tend to have high crural-index values, and vice versa. Nean- 0
derthals had a crural index similar to those of present-day 80 85
Lapps, who live above the Arctic Circle. Mean index of tibia/femur length (crural index)
This distinctive Neanderthal body shape may have been an adaptation to conserve
heat in a very cold environment. In cold climates, animals tend to be larger and have
shorter and thicker limbs than do members of the same species in warmer environ-
ments. This is because the rate of heat loss for any body is proportional to its surface
area, so any change that reduces the amount of surface area for a given volume will
conserve heat. The ratio of surface area to volume in animals can be reduced in two
ways: by increasing overall body size or by reducing the size of the limbs. In contem-
porary human populations there is a consistent relationship between climate and body
proportions. One way to compare body proportions is to calculate the crural index,
which is the ratio of the length of the shinbone (tibia) to the length of the thighbone
(femur). As Figure 12.29 shows, people in warm climates tend to have relatively long
limbs in proportion to their height. Neanderthals resemble modern peoples living
above the Arctic Circle.
The larynx (or voice box) of modern humans is much lower than in other
primates—an arrangement that allows us to produce the full range of vowels used in
modern human languages. On the basis of reconstructions of the Neanderthal vocal
tract, anatomist Jeffrey Laitman of Mount Sinai School of Medicine argues that the
Neanderthals had vocal tracts much like those of other primates and could not have
produced the full range of sounds necessary for modern speech. Laitman and his col-
leagues recorded the positions of several anatomical landmarks on the basicrania (sin-
gular basicranium; bottom of the skull) of humans and several other primates. In
infant humans, this basicranial hump is gentle, like those of other primates; but as
humans mature, the bottom of the cranium develops a pronounced upward indenta-
tion. Laitman believes that this relatively high hump helps make room for the elon-
gated human vocal tract. Australopithecine adults have apelike basicrania, but the
basicranium in Homo ergaster is intermediate between those in apes and modern
humans. Laitman argues that the initial shift in H. ergaster may have facilitated breath-
ing through the mouth during heavy aerobic exercise. The basicrania in Neanderthals
show a less humped profile than those of H. ergaster, suggesting that the Neanderthals’
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 333
capacity for language was more restricted than the language capacities of even their
own ancestors.
It is also possible that Laitman’s reconstructions are correct, but that grammati-
cally complex language still appeared long before modern humans evolved. It is pos-
sible for a creature with a high larynx to make all of the vowel sounds by using the
nasal cavity. In modern humans, such nasalized vowels take longer to produce and are
harder to understand. Thus, the lowering of the larynx might have served merely to
decrease the error rate of understanding a language that was otherwise very modern.
Second, as Harvard University linguist Steven Pinker points out in his book The Lan-
guage Instinct, “E lengeege weth e smell nember ef vewels cen remeen quete expres-
seve.” So, even if Laitman is correct, early hominins could have had fancy languages.
There is a lot of evidence that Neanderthals made Mode 3 tools and hunted
large game.
Careful study of Neanderthal skeletons indicates that Neanderthals didn’t live very
long. The human skeleton changes throughout the life cycle in characteristic ways, and
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 335
these changes can be used to estimate the age at which fossil hominins died. For exam-
ple, human skulls are made up of separate bones that fit together in a three-
dimensional jigsaw puzzle. When children are first born, these bones are still separate,
but later they fuse together, forming tight, wavy joints called sutures. As people age,
these sutures are slowly obliterated by bone growth. By assessing the degree to which
the sutures of fossil hominins have been obliterated, anthropologists can estimate how
old the individual was at death. Several other skeletal features can be used in similar
ways. All of these features tell the same story: Neanderthals died young. Few lived
beyond the age of 40 to 45 years.
Many of the older Neanderthals suffered disabling disease or injury. For example,
the skeleton of a Neanderthal man from La Chapelle-aux-Saints shows symptoms of
severe arthritis that probably affected his jaw, back, and hip. By the time this fellow
died, around the age of 45, he had also lost most of his teeth to gum disease. Another
individual, Shanidar 1 (from the Shanidar site in Iraq), suf-
fered a blow to his left temple that crushed the orbit (Figure
12.32). Anthropologists believe that his head injury probably
caused partial paralysis of the right side of his body, and this
in turn caused his right arm to wither and his right ankle to
become arthritic. Other Neanderthal specimens display bone
fractures, stab wounds, gum disease, withered limbs, lesions,
and deformities.
In some cases, Neanderthals survived for extended peri-
ods after injury or sickness. For example, Shanidar 1 lived
long enough for the bone surrounding his injury to heal.
Some anthropologists have proposed that these Nean-
derthals would have been unable to survive their physical
impairments—to provide themselves with food or to keep up
with the group—had they not received care from others. Many Neanderthal skeletons show signs of
FIGURE 12.32
Some researchers argue further that these fossils are evidence injury or illness. The orbit of the left eye of
of the origins of caretaking and compassion in our lineage. this individual was crushed, the right arm was withered, and the right
Katherine Dettwyler, an anthropologist at the University ankle was arthritic.
of Delaware, is skeptical of this claim. She points out that in
some contemporary societies, disabled individuals are able to support themselves, and
they do not necessarily receive compassionate treatment from others. In addition, non-
human primates sometimes survive despite permanent disabilities. At Gombe Stream
National Park, a male chimpanzee named Faben contracted polio, which left him com-
pletely paralyzed in one arm. Despite this impairment, he managed to feed himself, tra-
verse the steep slopes of the community’s home range, keep up with his companions,
and even climb trees (Figure 12.33).
Hominins living in Africa during the later Middle Pleistocene were more similar
to modern humans than were Neanderthals.
The fossil record in Africa indicates fairly clearly that African humans of this period
were not like the Neanderthals, although it is not entirely clear what they were like,
because the fossil record for this time in Africa is not nearly as good as it is in Europe.
A number of fossils dated to later in the Middle Pleistocene (300 to 200 kya) have robust
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 336
features similar to those of Homo heidelbergensis. Examples include the Florisbad cra-
nium found in South Africa (Figure 12.34) and the Ngaloba cranium (LH 18) from Lae-
toli, Tanzania. Like the Neanderthals, these African hominins show large cranial
volumes ranging from 1,370 to 1,510 cc. However, none of the African fossils shows
the complex of specialized features that are diagnostic of the European Neanderthals.
Although these fossils are variable and some are quite robust, many researchers believe
that they are more like modern humans than are the Neanderthals or earlier
African hominins.
About 190 kya, hominins belonging to our own species began to
appear in Africa, although some of the evidence is problematic. The oldest
fossils classified as Homo sapiens were excavated in 1963 at a site in south-
ern Ethiopia called Omo Kibish, where paleoanthropologists found most of
two fossil skulls and a number of other bone fragments. These skulls are
quite robust, with prominent browridges and large faces. However, one of
them has a number of modern features—most notably a high, rounded
braincase (Figure 12.35). Initially, these specimens could not be dated, but
they were recently dated to about 190 kya by new radiometric methods.
Several similar skulls were uncovered at Herto, another site in Ethiopia, and
are dated to 160 kya. Other more fragmentary fossil materials from other
sites in Africa also suggest that more modern hominins were present in
Africa between 200 and 100 kya. The archaeological record indicates that
Africans during this period developed more sophisticated technology and
This fossil skull, called
FIGURE 12.35 social behavior than did their contemporaries in Europe or Asia—a pattern
Omo Kibish 1, was found
in southern Ethiopia in 1963 and recently dated to that is consistent with the idea that this period saw the gradual accumula-
190 kya. It lacks the distinctive features of Nean- tion of the cognitive and behavioral characteristics that make modern
derthals: the face does not protrude, and the brain- humans so different from other hominins. We turn to this evidence in the
case is higher and shorter than that of Neanderthals. next chapter.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 337
The appearance of new tool technologies at 500 kya and 250 kya coincides neatly with
the timing of interglacial periods, consistent with the repeated replacement of Eurasian
populations by African populations.
You will notice that we avoided assigning a species name to the Neanderthals. This is
partly because there is a lot of disagreement about how to classify Middle Pleistocene
hominins. The disagreement stems from different ideas about the processes that shape
human evolution during this period. One school of thought, most closely associated
with University of Michigan anthropologist Milford Wolpoff, contends that hominins
in Africa and Eurasia formed a single interbreeding population throughout the Pleis-
tocene. The size of Africa and Eurasia limited gene flow and allowed regional differ-
ences to evolve, but there was always enough interbreeding to guarantee that all
hominins belonged to single species during this entire period. Thus, Wolpoff and oth-
ers prefer to include all hominins who lived during the Middle Pleistocene in a single
species, Homo sapiens, as shown in Figure 12.37a. As we will show in the next chap-
ter, this point of view has become increasingly difficult to sustain in the face of infor-
mation that has become available from molecular genetic studies in the last few years.
Other anthropologists believe that hominins split into several new species as
they migrated out of Africa and into Eurasia, although there are disagreements
about the details of phylogenetic history. Figure 12.37b and c illustrate two recent
hypotheses. According to G. Phillip Rightmire of Binghamton University, the spec-
imens that we have identified as Homo ergaster and Homo erectus are members
of a single species (which would be called Homo erectus because that was the
name first used) (Figure 12.37b). However, he draws a distinction between early
Middle Pleistocene hominins in Europe and Africa and those who came later. He
groups all of the early Middle Pleistocene hominins in Europe and Africa into one
species, which he labels H. heidelbergensis, and he lumps the hominins of the later
Middle Pleistocene in Europe and the Neanderthals into one species, H. nean-
derthalensis. Although it may seem peculiar to include specimens without the dis-
tinctive Neanderthal features in the species H. neanderthalensis, this name is used
because it has historical priority.
Richard Klein believes that hominin populations in Africa, western Eurasia, and
eastern Eurasia were genetically isolated from each other during most of the Pleis-
tocene and represent three distinct species (Figure 12.37c). In Africa, Homo ergaster
gradually evolved into Homo sapiens about 500 kya. Isolated in Asia, H. ergaster
evolved into Homo erectus early in the Pleistocene. Klein considers the development
of larger brains and more modern-looking skulls in the eastern Eurasia fossils, like
those found at Yingkou, to be the result of convergent evolution and includes them in
H. erectus. Once hominins reached Europe about 500 kya, they became isolated from
African and east Asian populations, and diverged to become H. neanderthalensis.
This kind of disagreement is to be expected because the timescales of change
become short as we come closer to the present. Instead of being interested in events that
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 339
Western Africa & East Asia & Western Africa & East Asia &
Eurasia Middle East Australasia Eurasia Levant Australasia
Homo sapiens
Homo
0.2 0.2 neander-
thalensis
0.4 0.4
1.0 1.0
Homo erectus
1.2 1.2
1.4 1.4
1.6 1.6
1.8 1.8
(a) (b)
0.6 during the Pleistocene. Illustrated here are three recent proposals: (a) Milford Wolpoff
holds that, beginning about 1.8 mya, there was only one human species. Because there
0.8 have been no speciation events since that time, all human fossils should be classified as
Homo Homo sapiens. (b) G. Phillip Rightmire believes that both African and Asian specimens
1.0 ergaster of Homo erectus should be classified as a single species. Approximately 800 kya, a
larger-brained species, Homo heidelbergensis, evolved in Africa and eventually spread
1.2
to Europe and perhaps east Asia. In Europe, Homo heidelbergensis gave rise to the
Neanderthals; in Africa, it gave rise to Homo sapiens. (c) Richard Klein argues that
1.4
Homo ergaster evolved in Africa about 1.8 mya and soon spread to Asia, where it
1.6 differentiated to become a second species, Homo erectus. About 500 kya, Homo
ergaster spread to western Eurasia, where it evolved into Homo neanderthalensis. In
1.8 Africa, Homo ergaster evolved into Homo sapiens at about the same time.
(c)
took place over millions of years, we are now interested in events that took place in just
a hundred thousand years. But this may be roughly how long it took for two species to
diverge during the process of allopatric speciation (see Chapter 4). Thus, as hominins
spread out across the globe and encountered new habitats, regional populations may
have become isolated and experienced selection—conditions that would eventually lead
to speciation. The rapidly fluctuating climates of the Pleistocene, however, caused the
ranges of hominins and other creatures to shift as well. As their ranges expanded and
contracted, some populations may have become extinct, some may have become fully
isolated and even more specialized, and others may have merged together. This com-
plex set of possibilities is very difficult for paleontologists to unravel.
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 340
FURTHER READING
Conroy, G. 2005. Reconstructing Human Origins: A Modern Synthesis. 2nd ed. New
York: Norton.
Klein, R. G. 1999. The Human Career: Human Biological and Cultural Origins. 2nd
ed. Chicago: University of Chicago Press.
McBrearty, S., and A. S. Brooks. 2000. The revolution that wasn’t: a new interpreta-
tion of the origin of modern human behavior. Journal of Human Evolution
39:453–563.
Rightmire, G. P. 1998. Human evolution in the Middle Pleistocene: the role of Homo
heidelbergensis. Evolutionary Anthropology 6:218–227.
Stringer, C., and C. Gamble. 1993. In Search of the Neanderthals: Solving the Puzzle
of Human Origins. New York: Thames and Hudson.
Walker, A., and P. Shipman. 1996. The Wisdom of the Bones: In Search of Human
Origins. New York: Knopf.
1. Which derived features are shared by modern humans and Homo ergaster?
Which derived features are unique to H. ergaster?
2. What does the specimen KNM-WT 15000 tell us about H. ergaster?
3. What is a hand ax, and what did H. ergaster use hand axes for? How do we
know?
4. What evidence suggests that H. ergaster controlled fire?
5. What evidence suggests that H. ergaster ate significant amounts of meat? Is the
evidence for H. ergaster carnivory better than that for the Oldowan hominins?
Explain.
6. What are the main differences between H. ergaster and H. erectus? How would
you explain the evolution of these differences?
7. Using present-day examples, describe the variation in climate during the Middle
Pleistocene. Why is this variation important for understanding human evolution?
8. How is H. heidelbergensis different from H. erectus?
9. What important technological transition occurred about 300 kya? Why was it
important?
10. Explain why the fossils found at Sima de los Huesos are significant.
11. What is the crural index? What does it measure? How does the crural index of the
Neanderthals differ from that of modern tropical peoples?
12. How did the Neanderthals differ from their contemporaries in Africa and eastern
Asia?
13. What evidence suggests that western Eurasia was subjected to repeated invasions
from Africa during the Middle Pleistocene?
12_Hominin to Homo [4p]:How Humans Evolved [5e] 10/10/08 12:13 PM Page 341
KEY TERMS
Chapter 13
Morphological Features of Modern
Homo sapiens
Genetic Features of Modern Homo sapiens
Archaeological Evidence for Modern
Human Behavior
Upper Paleolithic Technology and Culture
The Origin and Spread of Modern Humans
Genetic Data
Evidence from Fossils and Tool Kits
Modern Human Behavior: Revolution
or Evolution?
The African Archaeological Record
during the Later Pleistocene
How Modern Human Behavior Evolved
B
etween 40 and 30 kya, the fossil and archaeological record in Europe under-
goes a striking change: the Neanderthals disappeared and were replaced by
essentially modern people. These people looked much like people in the world
today, with high foreheads, sharp chins, and less robust physiques. The archaeological
record suggests that they also behaved like modern people, using sophisticated tools,
trading over long distances, and making jewelry and art. Anatomically and behaviorally
similar people appeared in Australia around the same time. There is evidence that sim-
ilar kinds of peoples also lived in Africa at this time, and most anthropologists think
that modern people first appeared in Africa well before they appeared in Europe.
342
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 343
In this chapter, we describe in more detail the fossil and archaeological record for
these early modern people. Then we consider how they evolved. You will see that evi-
dence from the fossil record and from molecular genetic studies of DNA extracted
from fossils gives us a fairly clear picture of where and when modern human mor-
phology evolved. These data tell us that modern-looking, or anatomically modern,
humans evolved in Africa between 200 and 100 kya and then migrated from Africa
about 60 kya and eventually spread throughout the world. We are less certain when
and where modern human behavior evolved. According to one view, anatomically
modern people inhabiting Africa 100 kya still lived much like the more robust
hominins in Europe and Asia. Then, about 60 kya, an important cognitive innovation
arose that allowed people to develop more complex technology and social lives, pro-
ducing what anthropologists call the “human revolution.” Adherents of a second view
argue that the components of modern human behavior and technology gradually
evolved in Africa along with modern human morphology over a period of about
200,000 years. There was no human revolution in Europe or elsewhere in the world;
what seems like a revolutionary change in Europe simply reflects the migration of mod-
ern people from Africa to Europe and the existence of an extremely rich archaeolog-
ical record that is heavily biased in favor of European sites.
The first fossils of modern humans in Europe were found in 1868 by railroad work-
ers in southwestern France at a site known as the Cro-Magnon rock shelter. This site
eventually yielded bones from at least five different individuals who lived about 30
kya. Since this time, many other sites in Europe have yielded fossils of modern humans
or their artifacts. These creatures are classified by anthropologists as Homo sapiens
because they share a number of important derived features with contemporary
humans that were not present in previous hominins (Figure 13.1):
Vertical
forehead
Back of
cranium
rounded The skulls of
FIGURE 13.1
modern humans
Small have higher, rounder crania and smaller
face faces than earlier hominins had, as illus-
trated by this skull from a man who lived
about 25 kya near the Don River in
Russia. (Figure courtesy of Richard
Chin Klein.)
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 344
344 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
Strong
Occipital browridge
bun
Projecting
Rather large, midface
wide nose
Shanidar 1 La Ferrassie 1
Rather short,
round, high High forehead
braincase
Rounded
occipital Smaller
Flatter
nose
midface
Chin
Qafzeh 9 ˘
Predmostí 3
• Small, flat face with protruding chin. These people had smaller faces and smaller
teeth than earlier hominins had. The face is flat and is tucked under the the brain-
case. The lower jaw had a jutting chin for the first time (Figure 13.2). Some
anthropologists believe that the smaller face and teeth were favored by natural
selection because these people did not use their teeth as tools as much as earlier
people had. There is no agreement about the functional significance of the chin.
• Rounded skull. Like modern people, these people had high foreheads, a distinc-
tive rounded back of the cranium, and greatly reduced browridges (see Figure
13.2).
• Cranial capacity of at least 1,350 cc. Cranial capacity varies to some extent across
populations but is generally at least 1,350 cc. This value is smaller than the value
for Neanderthals but greater than the value for other hominins in the late Mid-
dle Pleistocene.
• Less robust postcranial skeleton. The skeleton of these people was much less
robust than Neanderthal skeletons. These people had longer limbs with thinner-
walled bones; longer, more lightly built hands; shorter, thicker pubic bones; and
distinctive shoulder blades. Although these people were less robust than Nean-
derthals, they were still more heavily built than any contemporary human popu-
lation. Erik Trinkaus of Washington University in St. Louis suggests that these
people relied less on body strength and more on elaborate tools and other tech-
nological innovations to do their work.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 345
• Relatively long limbs and short trunks. The body proportions of these creatures
were similar to those of peoples who live in warm climates and may reflect the
African origins of these people.
In the past, humans were distinguished from other species mainly on the basis of mor-
phology, and fossils were our only source of information about human evolutionary
change. But today, molecular genetics provides us with another source of information
about our evolutionary history. In this section, we summarize what is known about
how modern humans differ genetically from our closest relative, the chimpanzee.
The genome of an organism is all of the genetic information carried on chromo-
somes in that species. For many years, our knowledge of the genome was indirect: we
could study the protein products of genes, and we could use specialized molecular tech-
niques to identify particular DNA segments called “genetic markers.” Lately, however,
it has become possible to sequence large chunks of the genome. In 2002, a first cut at
sequencing the entire human genome was announced with great fanfare, and as of this
writing the complete genomes of several people have been sequenced. More quietly,
geneticists have worked on sequencing the genomes of many other organisms, includ-
ing yeast, fruit flies, mice, cats, dogs, and rhesus macaques. Of particular interest to
us is the genome of one of our closest relatives, the chimpanzee. This vast trove of
genetic information, published in 2006, gives us new insights about the kinds of evo-
lutionary changes that have occurred in the human lineage.
By aligning the human and chimpanzee genomes and comparing the sequences
nucleotide by nucleotide, geneticists have been able to measure the magnitude of the
genetic differences between the two species. Humans and chimpanzees differ by just
1.06%. This means that in about 1.06% of the nucleotides, all chimpanzees have one
nucleotide and all humans have a different nucleotide. This sounds like a minuscule
difference, but remember that there are 3 billion bases in the human genome. A 1.06%
difference represents differences in about 30 million nucleotides. There have also been
approximately 5 million insertions and deletions of bits of DNA in or out of the
human genome or the chimpanzee genome. Most of these insertions and deletions
involve a small number of nucleotides, and most involve repetitive sequences or trans-
posable elements, copies of of DNA segments from one part a genome that have been
inserted somewhere else. Insertions and deletions contribute another 3% to the over-
all difference between the genomes.
Many people are puzzled by the small genetic difference between humans and
chimpanzees. How could such a puny genetic difference produce the sizable phenotypic
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 346
346 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
differences between humans and chimpanzees? The answer is that small sequence dif-
ferences can lead to big differences in phenotype because the percentage of DNA that
differs is not the same as the percentage of genes that differ. To understand why this
is true, consider two extreme possibilities. If DNA differences were distributed evenly
across all of the genes in the genome, then even a 1% difference between two species
would cause every gene in the two species to differ. On the other hand, if the differ-
ences were clustered in certain parts of the genome, then a smaller proportion of the
genes would differ. So, clearly it is important to know something about the pattern of
differences between chimpanzees and humans, as well as the overall magnitude of
differences.
This is where we benefit from having sequenced the complete genomes of humans
and chimpanzees. Protein coding genes (DNA sequences that code for proteins) can
be identified in a DNA sequence by the “start” and “stop” codons that mark the begin-
ning and end of each coding sequence. Using this approach, geneticists identified
13,454 homologous protein coding genes in chimpanzees and in humans. Only 29%
of these genes had exactly the same amino acid sequences. Among those that differed,
the median number of base substitutions is two. Thus, even though the DNA
sequences of chimpanzees and humans differ by only a small percentage, 71% of the
proteins produced by their genes differ.
Only a small fraction of protein coding genes shows evidence of selection since
the divergence of human and chimpanzee lineages.
For example, the consortium of scientists that was responsible for sequencing the chim-
panzee genome estimated that about 4.4% of the genes showed signs of positive selec-
tion. A more recent study by another research group places the estimate at 2.7% These
estimates represent upper bounds on the actual percentage of positively selected genes
because a substantial number of cases would be expected to occur by chance alone.
Thus, even though most coding sequences differ, only a small fraction of these differ-
ences seem to be functional. Again, this result is surprising given the magnitude of the
phenotypic differences between humans and chimpanzees.
Two other aspects of these studies are hard to reconcile with what we know about
human evolution. First, the human lineage has undergone extensive change, especially
to the nervous system and the brain. However, the positively selected genes that have
been identified mainly code for proteins that are related to the immune system, the
reproductive system, and the sense of smell. These are systems that seem to be evolv-
ing rapidly in many different mammalian lineages and don’t relate to the changes pecu-
liar to the human lineage. Second, humans are more different from other primates than
chimpanzees are, so it seems that there must have been more rapid evolution in the
human lineage than in the chimpanzee lineage after the two lineages split. However,
the sequence data show just the opposite. It has been estimated that 60% of the 387
positively selected genes have occurred in the chimpanzee lineage and that 40% have
occurred in the human lineage.
How can the apparent lack of genetic change be reconciled with the substantial
amount of phenotypic change? One possibility is that measuring selection by com-
paring synonymous and nonsynonymous changes underestimates the amount of
change due to selection. This method assumes that the amount of evolution of a pro-
tein coding gene is proportional to the number of nonsynonymous DNA bases that dif-
fer between two species. However, sometimes the change of one or two base pairs in
a DNA sequence can strongly affect phenotype. For example, we will see in the next
chapter that the FOXP2 gene in humans has a major impact on speech, although the
human and chimpanzee versions differ by only two substitutions. Such small sequence
changes are not usually detected by counting synonymous and nonsynonymous sub-
stitutions (and the FOXP2 gene was not identified as a positively selected gene).
A second possibility is that most of the evolutionary changes are not the result of
changes in protein coding genes, but changes in regulatory genes. Unfortunately, we
are not yet able to identify regulatory genes from DNA sequence data alone, so geneti-
cists can’t study the changes in regulatory genes in the same way that they study
changes in structural genes.
A team led by Katherine Pollard of the University of Californa, San Francisco used
a different method to identify regions that have experienced significant positive selec-
tion after human and chimpanzee lineages diverged. They searched the genomes of the
mouse, rat, and chimpanzee to find DNA sequences that were at least 100 base pairs
in length and were at least 96% identical in all three taxa. Because the two rodent
species are separated from chimpanzees by about 70 million years of independent evo-
lution, Pollard and her colleagues reasoned that these sequences must be subject to
strong negative selection—selection that favored the observed stable sequence over
mutants that arose during these millions of years. They found about 35,000 negatively
selected sequences. Then for each of these regions, they compared the rate of change
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 348
348 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
in the human lineage with the average rates of change in 12 other vertebrate species
(not including the chimpanzee, rat, or mouse). The rates of change were significantly
greater in the human lineage than in other lineages in 202 of these regions. The inves-
tigators ranked the regions by the rate of change and then assigned each region a label
based on the ranking. Thus, HAR1 was the fastest of the fast, and HAR202 was the
slowest of the fast; HAR stands for “highly accelerated region.” Almost all of these
HAR segments are in noncoding regions. Even though these segments do not encode
the structure of proteins and, for the most part, we don’t know what they do, we do
know that they have evolved very rapidly during human evolution. This suggests that
they have been shaped by natural selection, not genetic drift.
The fastest-changing region, HAR1, provides one example of how this might
work. HAR1 is a 118-base-pair sequence on chromosome 20. In other vertebrates,
HAR1 is extremely conservative, showing only 2 base-pair changes between the
chicken and the chimpanzee. If this rate of change had been continued in the human
lineage, there would be only a 25% chance of having even one difference between the
sequence in chimpanzees and modern humans. However, there have been 18 base-pair
changes, nearly an 80-fold increase in the rate of evolution. The HAR1 segment codes
for an RNA molecule that folds itself into a stable structure. These kinds of RNA mol-
ecules often work with proteins to regulate gene expression, and it seems likely that
this is what HAR1 does. Studies by Pollard and her coworkers have shown that HAR1
is expressed exclusively in the brain, especially in the developing brain. There, it is
associated with a protein called reelin, which is associated with the development of the
layered structure that is characteristic of human brains but is not seen in other species.
Thus, it seems likely that the rapid change in HAR1 during human evolution is related
to the rapid evolution of the larger and more complex human brain.
Future work will amplify our understanding of how and when genetic changes
arose in the human lineage.
Studies of the genetic changes that have occurred in the human lineage are still in
their infancy. The technologies that allow us to study these changes are still very new,
and it will take time for scientists to process and interpret the meaning of the data that
these technologies produce. But there is no doubt that these data will provide us with
information that will be crucial to our understanding of how and when particular
traits in humans evolved.
Modern human behavior is much more complex and much more variable than the
behavior of earlier hominins. Remember that Homo ergaster and Homo heidelber-
gensis used Acheulean tools throughout Africa and western Eurasia for more than a
million years. Thus, the same tools, and presumably more or less the same adaptive
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 349
strategies, were used over a wide range of habitats for a very long period of time. In
contrast, present-day foragers use a vast range of different, highly specialized tools and
techniques to adapt to a very large range of environments. They also engage in elab-
orate and varied symbolic, artistic, and religious behavior that is unparalleled among
other creatures. The extraordinary geographic range and sophistication of modern
humans is due partly to our cognitive abilities. Individuals can solve problems that
would completely stump other creatures (see Chapter 8). However, being smarter than
the average bear (or primate) is only part of our secret. Our other trick is having the
ability to accumulate and transmit complex adaptive and symbolic behavior over suc-
cessive generations. People rarely solve difficult problems on their own. Even the most
brilliant of us couldn’t construct a seaworthy kayak or invent perspective drawing
from scratch. Instead, we gain skills, knowledge, and techniques from watching and
being instructed by others. The variety and sophistication of modern human behav-
ior is a product of our ability to acquire information in this way (see Chapter 15).
The archaeological record provides considerable evidence that the first Homo
sapiens in Europe had achieved this kind of complex and varied behavior:
• Ecological range. They were able to occupy a difficult, cold, dry habitat and
extended their range farther north and east than previous hominins did.
• Technology. They assembled more sophisticated and more highly standardized
tools made from a wider variety of materials, including antler, ivory, and bone.
They also constructed elaborate shelters.
• Social organization. They used raw materials that came from sources hundreds of
kilometers away for toolmaking, suggesting that they had long-distance exchange
networks.
• Symbolic expression. They created art and ornamentation, performed ritual buri-
als, and practiced other forms of symbolic behavior.
As we will explain more fully later in this chapter, the first modern humans in
western Eurasia created a number of different tool industries in different parts of their
range. Archaeologists refer to these tool industries collectively as Upper Paleolithic
industries to distinguish them from the earlier tool industries associated with Nean-
derthals and other earlier hominins. The earliest Upper Paleolithic tools appear in the
Near East about 45 kya, and they disappear from the archaeological record about 10
kya. Archaeologists refer to this period in western Eurasia as the Upper Paleolithic
period and the people who made the tools as Upper Paleolithic peoples.
The fossil and archaeological record in Australia is also quite good. Numerous
well-dated archaeological sites indicate that modern Homo sapiens entered Australia
at least 40 kya. Several sites may be older, but there is controversy about the dates asso-
ciated with these sites. What is certain is that people occupied the entire continent by
30 kya. The most complete early site is in southeastern Australia at Lake Mungo. The
finds here include three hominin skulls, hearths, and ovens that have been dated to 32
kya using carbon-14 dating. The skull specimens are fully modern and well within the
range of variation of contemporary Australian aborigines.
The earliest Australians seem to have practiced some of the same kinds of cultural
behaviors as the Upper Paleolithic Europeans did. Some of the tools found at these
early sites were made from bone, there are cave paintings dated to 17 kya, and there
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 350
350 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
is evidence of both ceremonial burials and cremation. About 15 kya, Australians seem
to have been the first people to use polished stone tools, which are made by grinding
rather than flaking. In other parts of the world, polished tools do not appear in the
record until agriculture is introduced, about 7 kya.
The fact that people got to Australia at all is evidence of their technological sophis-
tication. Much of the world’s water was tied up in continental glaciers 40 kya, so New
Guinea, Australia, and Tasmania formed a single continent that geologists call Sahul.
However, there was still at least 100 km (62 miles) of open ocean lying between Asia
and Sahul. This gap was wide enough to prevent nearly all movement of terrestrial
mammals across it and, thus, to preserve the unique, largely marsupial fauna of Aus-
tralia and New Guinea. The colonization of Sahul cannot be dismissed as a single lucky
event, because people also crossed another 100 km of ocean to reach the nearby
islands of New Britain and New Ireland at about the same time. The people who first
settled these islands must have been able to build seaworthy boats, a sophis-
ticated form of technology.
Upper Paleolithic peoples manufactured blade tools, which made efficient use
of stone resources.
During this period, humans shifted from the manufacture of round flakes to the man-
ufacture of blade tools. Blades are stone flakes that look like modern knife blades: they
are long, thin, and flat, and have a sharp edge. Blades have a longer cutting edge than
flakes do, so blade technology made more efficient use of raw materials than older tool
technologies did. However, there was a cost: although blades made more efficient use
of materials, they also took more time to manufacture, requiring more preparation and
more finishing strokes. Archaeologists label tool kits that emphasize blades as Mode 4
technologies.
Upper Paleolithic peoples made many more kinds of tools than earlier hominins
had made. Chisels, various types of scrapers, a number of different kinds of points,
knives, burins (pointed tools used for engraving), drills, borers, and throwing sticks
are just some of the items from the Upper Paleolithic tool kit.
Even more striking is the fact that the different tool types have distinctive, stereo-
typed shapes. It is as if the Upper Paleolithic toolmakers had a sheaf of engineering
drawings on which they recorded their plans for various tools. When the toolmaker
needed a new 10-cm (4-in.) burin, for example, she would consult the plan and pro-
duce one just like all the other 10-cm burins. Of course, these
toolmakers didn’t really use drawings as modern engineers
do, but the fact that we find standardized tools suggests that
they carried these plans in their minds. The final shape of
Upper Paleolithic tools was not determined by the shape of
the raw material; instead, the toolmakers seem to have had
a mental model of what the tool was supposed to look like,
and they imposed that form on the stone by careful flaking.
The craft approach to toolmaking reached a peak in the
Solutrean tool tradition, which predominated in southern
France between 21 and 16.5 kya (Figure 13.4). Solutrean
toolmakers crafted exquisite blades shaped like laurel leaves
that were sometimes 28 cm (about 1 ft) long, 1 cm (about 1⁄2
in.) thick, and perfectly symmetrical.
Upper Paleolithic people were also the first to shape
tools from bones, antlers, and teeth. Earlier hominins had
made limited use of bone, but Upper Paleolithic people
transformed bone, ivory, and antler into barbed spear points, Long, thin, and delicate blades characterized
FIGURE 13.4
awls, sewing needles, and beads. the Solutrean tool tradition.
The first Upper Paleolithic industry in Europe, the Aurignacian, was widespread
in Europe by 35 kya (Figure 13.5). It is characterized by certain types of large blades,
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 352
352 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
burins, and bone points. About 27 kya, the Aurignacian was replaced in southern
France by a new tool kit, the Gravettian, in which small, parallel-sided blades pre-
dominated and bone points were replaced by bone awls. Then about 21 kya, the
Solutrean, with its beautiful leaf-shaped points, developed in the region. And 16 kya,
the Solutrean gave way to the Magdalenian, a tool kit dominated by carved, decorated
bone and antler points. Other parts of Europe are characterized by different sequences
of tool complexes; so after the Aurignacian, each region typically has a distinctive
material culture. Over roughly 25,000 years of the Upper Paleolithic in Europe, there
were dozens of distinctive tool kits emphasizing Mode 4 tools. This diversity stands
in striking contrast to the Acheulean tool kit, which remained unchanged throughout
more than half of the Old World for over a million years.
Stones and other raw materials for toolmaking were often transported hundreds
of kilometers from their place of origin.
Modern humans exploited a wider range of prey species than did the
Neanderthals, but the subsistence economies of the two populations
were similar.
The Upper Paleolithic spanned the depths of the last ice age. The area in Europe
that contains the richest archaeological sites was a cold, dry grassland that supported
large populations of a diverse assemblage of large herbivores, including reindeer (called
“caribou” in North America), horse, mammoth, bison, woolly rhinoceros, and a vari-
ety of predators, such as cave bears and wolves.
Bones found at Upper Paleolithic sites indicate that large herbivores played an
important role in the diets of Upper Paleolithic peoples. Everywhere, Upper Paleolithic
peoples hunted herbivores living in large herds, fished, and hunted birds. In some
places, they concentrated on a single species—reindeer in France, red deer in Spain
(Figure 13.6), bison in southern Russia, and mammoths farther north and east. In
some areas, such as the southern coast of France, rich salmon runs may have been an
important source of food. There are also places where these peoples harvested several
different kinds of animals.
Modern humans probably made extensive use of plant foods, but few sites
preserve the remains from vegetation.
All modern foraging peoples rely on plant foods that they gather as well as ani-
mals that they hunt. It is likely that earlier hominins made use of plant foods as well,
but the remains of these foods have not been preserved. Unusual preservation condi-
tions at several sites provide glimpses into this part of the subsistence economy of
Upper Paleolithic peoples. Dani Nadel of the Hebrew University of Jerusalem and his
colleagues have excavated a site on the shores of the Sea of Galilee that is dated to 19
Wolf
Fox
Species of bones in Altamira assemblage
Brown
bear
Bison
Roe deer
Horse
The bones
FIGURE 13.6
of red deer
Red deer
dominate the assemblage at Altamira
in northern Spain, as illustrated by
0 10 20 30 40 50 60 these data, which come from a Mag-
Percentage of sample dalenian site dated to 15 to 13 kya.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:12 PM Page 354
354 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
kya. This site, which is called Ohalo II, contains more than 90,000 specimens of plants
from 142 different plant taxa. These include wild varieties of barley, wheat, acorns, pis-
tachios, olives, raspberries, figs, and grapes. Starch grains from barley have been found
on a grinding stone, indicating that wild cereal grains were processed to make food.
The peoples of the Upper Paleolithic developed more complex forms of shelter
and clothing than the Neanderthals had.
In what is now western and central Europe, Russia, and the Ukraine, the remains
of small villages have been found. Living on a frigid, treeless plain, Upper Paleolithic
peoples either hunted or scavenged mammoth and used the hairy beasts for food, shel-
ter, and warmth. At the site of Předmostí in the Czech Republic, the remains of almost
100,000 mammoths have been discovered. These peoples constructed huts by arrang-
ing mammoth bones in an interlocking pattern and then draping them with skins.
[Temperatures of about ⫺45°C (⫺50°F) outside provided a strong incentive for
chinking the cracks!] The huge quantities of bone ash that have been found at these
sites indicate that the mammoth bones were also used for fuel. A site about 470 km
(about 300 miles) southeast of Moscow contains the remains of even larger shelters.
They were built around a pit about a meter deep and were covered with hides sup-
ported by mammoth bones. Some of these huts had many hearths, suggesting that a
number of families may have lived together (Figure 13.7).
Modern foragers living in warm places often construct simple brush huts, using
large branches to support the roof and smaller twigs and grasses to form the walls and
Hearths
N Large bones
Mammoth teeth
Mammoth crania
Mammoth crania
and scapulae
cover the roof. Grasses may be gathered and spread on the floor and used as bedding.
At the Ohalo II site, Nadel and his colleagues have found kidney-shaped depressions
that are about 5 by 13 sq m in area. These depressions contain substantial pieces of
tree branches that might have held up the roof of a hut, as well as smaller twigs and
stems that might have been part of the roof of the shelter. In one of the depressions,
archaeologists have found grass stems that were loosely woven together around a mass
of ash. It looks like the grasses may have formed a mat around a fire that people sat
on as they cooked and slept on at night. If this interpretation is correct, then this site
represents the oldest evidence for the construction brush huts.
Several lines of evidence indicate that Upper Paleolithic peoples living in glacial
Europe manufactured fur clothing. First, when modern hunters skin a fur-bearing ani-
mal, they usually leave the feet attached to the pelt and discard the rest of the carcass.
Numerous skeletons of foxes and wolves that are complete except for their feet have
been found at several Upper Paleolithic sites in Russia and the Ukraine, suggesting that
early modern humans kept warm in sumptuous fur coats. Second, bone awls and bone
needles are common at Upper Paleolithic sites, so sewing may have been a common
activity. Finally, three individuals at a burial site in Russia seem to have been buried
in caps, shirts, pants, and shoes lavishly decorated with beads.
Upper Paleolithic peoples may have begun the process of domesticating dogs.
All domesticated dogs are the descendants of gray wolves, and mitochondrial
DNA evidence inicates that dogs and wolves diverged before 100 kya somewhere in
Eurasia or eastern Asia. Dogs have shorter and broader muzzles than wolves, and
skulls that bear these characteristics have been found at sites in Russia dated to 17 to
13 kya. By analogy with modern foragers, Upper Paleolithic peoples may have used
dogs to protect their camps against dangerous predators, as hunting partners, and to
carry or pull loads.
Upper Paleolithic peoples were better able to cope with their environment than
the Neanderthals were.
The richness of the fossil and archaeological record in Europe provides a detailed
comparison of Neanderthal and Upper Paleolithic peoples. Three kinds of data sug-
gest that Upper Paleolithic peoples were better adapted to their environment than
Neanderthals were:
356 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
lived much longer than the Neanderthals. About a third of their sample of 113
Neanderthals reached twice the age of first reproduction, which is about 30 years
of age in modern hunter-gatherer populations. In contrast, two-thirds of a sam-
ple of 74 Upper Paleolithic fossil individuals reached that age. The fact that
Upper Paleolithic populations included a substantial fraction of older individu-
als, while Neanderthals were dominated by the young, may have allowed Upper
Paleolithic peoples to retain and transmit more complex cultural knowledge than
Neanderthals could.
3. Upper Paleolithic peoples were less likely to suffer serious injury or disease than
were Neanderthals. In sharp contrast to the Neanderthals, the skeletons of Upper
Paleolithic people rarely show evidence of injury or disease. Among the few
injuries that do show up in the fossil record, there is a child buried with a stone
projectile point embedded in its spine, and a young man with a projectile point
in his abdomen and a healed bone fracture on his right forearm. Evidence of dis-
ease is slightly more prevalent than evidence of injury among the remains of
Upper Paleolithic peoples; affected specimens include a young woman who prob-
ably died as the result of an abscessed tooth, and a child whose skull seems to
have been deformed by hydrocephaly (a condition in which fluid accumulates in
Diadem
Necklace
Bracelet
Limestone
Retouched slabs
blade
Bone Bird
"button" figurine
Bone
point
Borer Blade
the cranial cavity and the brain atrophies). Nonetheless, there is still less evidence
for disease among these peoples than among Neanderthals.
There is good evidence for ritual burials during the Upper Paleolithic period.
Like the Neanderthals, Upper Paleolithic peoples frequently buried their dead.
Upper Paleolithic sites provide the first unambiguous evidence of both multiple buri-
als and burials outside of caves. Unlike the Neanderthals, Upper Paleolithic burials
appear to have been accompanied by ritual. The people frequently buried their dead
with tools, ornaments, and other objects that suggest they had some concept of life
after death. Figure 13.8 diagrams the grave of a child who died about 15 kya at the
Siberian site of Mal’ta. The child was buried with a number of items, including a neck-
lace, a crown (diadem), a figurine of a bird, a bone point, and a number of stone tools.
Upper Paleolithic peoples were skilled artisans, sculpting statues of animals and
humans and creating sophisticated cave paintings.
It is their art that distinguishes Upper Paleolithic peoples most dramatically from
the hominins who preceded them. They engraved decorations on their bone and antler
tools and weapons, and they sculpted statues of animals and female figures (Figure
13.9). The female statues are generally believed to be fertility figures because they usu-
ally emphasize female sexual characteristics. Upper Paleolithic peoples also adorned
themselves with beads, necklaces, pendants, and bracelets, and may have decorated
their clothing with beads.
Small animal
FIGURE 13.9
figures were
carved from mammoth tusks during the
Aurignacian period.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 358
358 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
The new evidence indicates that most, and perhaps all, of the genes that gave rise to
modern human morphology and behavior evolved in an African population sometime
between 200 and 100 kya. People carrying these genes subsequently spread through-
out Africa and differentiated into a number of morphologically modern but genetically
variable populations. Then about 60 kya, people from one of these regional popula-
tions left Africa and spread across the world, replacing other hominin populations with
little gene flow between them.
In this chapter, we will describe the genetic, paleontological, and archaeological
evidence related to the origin and spread of modern humans. As you will see,
researchers working in a diverse set of academic disciplines have brought a wide range
of methods to bear on this problem.
Genetic Data
The patterns of genetic variation within living people and genetic material extracted
from fossils tell us a lot about the history of human populations. We know that:
We do not know exactly what happened before the expansion 60 kya. It is not clear
whether there was much gene flow between African and Eurasian populations during
this period (Figure 13.12). All modern human genes may be inherited from a single
African population of about 10,000 adults, or some genes may have arisen in Asia
between 2 million and 100,000 years ago, diffused back to Africa, and then exited
again 60 kya.
Much of this information about genetic variation comes from genes carried on
the Y chromosome or on mitochondria.
Up until this point, we have focused on “plain vanilla” genes carried on regular
chromosomes in the nucleus. However, much of the information about human evo-
lution has come from more exotic genes carried on the Y chromosome and on
mitochondria.
The X and Y chromosomes determine sex in humans—females carry two X chro-
mosomes, and males carry one X and one Y chromosome. The X is a fairly normal
chromosome, but the Y is just plain weird. It is inherited by males from their fathers
and carries very few genes. It mostly consists of long stretches of simple repeated
sequences, and 95% of the material on the Y chromosome does not recombine with
material on the X chromosome.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 360
360 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
? ?
? ?
This is a diagram of the human phy-
FIGURE 13.12
logeny supported by gene trees con-
structed for mtDNA, NRY, and a number of genes carried on
ordinary chromosomes. The genetic data support the idea that all
modern humans are descended from a population that emerged
from Africa about 60 kya and expanded through a series of
stepped bottlenecks. However, it is not clear whether there was
gene flow from Eurasian populations during earlier parts of the
Middle Pleistocene.
The human species is less genetically variable than are other species.
Biologists have devised a number of ways to measure the amount of genetic vari-
ation within a population. For example, one important measure of genetic variation
is the average number of genetic differences per base pair (nucleotide) in a DNA
sequence. To compute this measure, geneticists collect tissue samples from members
of a population and assess the same part of the DNA sequence of each individual’s
mitochondria. Then, for each pair of individuals in the sample, they check to see
whether DNA at each position is the same or different. They repeat this for all pairs
of individuals in the sample and then compute the average number of differences per
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 361
0.125 0.125
0.125 0.125
0.1 0.1
Substitutions 0.1 Substitutions 0.1
0.075 0.075
per 0.075 per 0.075
0.05 0.05
nucleotide 0.05 0.025 nucleotide 0.05 0.025
0.025 0 0.025 0
0 verus 0 Africa
schweinfurthi troglodytes Europe Asia
troglodytes Asia
schweinfurthi Europe
verus Africa
Humans are less genetically variable than chimpanzees. (a) Plotted here is
FIGURE 13.13
the mean number of differences per nucleotide in one region of mtDNA for
three geographically separated chimpanzee populations: Pan troglodytes schweinfurthi (from East
Africa), P. t. troglodytes (from central Africa), and P. t. verus (from West Africa). (b) Here, the same
data are plotted for three geographically separated human populations (in Asia, Europe, and Africa). In
both graphs, each red bar at the intersection of a row and a column gives data for comparing the popu-
lation(s) identified at the ends of that row and column. Thus, the height of each red bar represents the
mean number of differences between pairs of individuals drawn from the same population. For example,
each pair of chimpanzees belonging to the verus population differs at about 7.5% of the nucleotides
sampled. Each pair of individuals drawn from the most variable human population, Africans, differs at
only about 2.5% of the nucleotides assessed. The height of the blue bars represents the mean number
of differences between pairs of populations. On average, an individual drawn from the verus population
differs from an individual drawn from the schweinfurthi population at about 13% of the nucleotides,
while Africans and Europeans, the most different pair of human populations, differ at less than 0.3% of
the nucleotides. Thus, the two graphs show that there is more variation within each of the chimpanzee
populations than within any of the human populations, and human populations on the average are more
similar to one another than chimpanzee populations are.
nucleotide across the sample. As populations become more diverse genetically, the
average number of genetic differences per nucleotide increases.
Humans are less genetically variable than chimpanzees. Figure 13.13 plots the
amount of mtDNA genetic variation within and between three geographically sepa-
rated human populations—Africans, Asians, and Europeans—and the amount of
genetic variation within and between chimpanzee populations in eastern, central, and
western Africa. The bars on the diagonal (red) show that the amount of variation
within each of these human populations is much lower than that within any of the
chimpanzee populations. These data tell us that any two humans taken at random
from a single population are much more similar to one another than are any two
chimpanzees taken at random from a single population. The off-diagonal bars (blue)
plot the average differences among human populations and among chimpanzee
populations.
The small amount of genetic variation indicates that the human population is
descended from a population of about 10,000 people.
The data on genetic distance suggest that much of the human genome evolved
mainly under the influence of drift and mutation. Remember that mutation introduces
new genes at a very low rate and that genetic drift eliminates genetic variation at a rate
that depends on the size of the population. In large populations, drift removes genetic
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 362
362 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
variants slowly, but in small populations, drift eliminates variation more quickly. The
effects of mutation and genetic drift will eventually balance (Figure 13.14) as the
amount of variation reaches an equilibrium. The amount of variation (m) at equilib-
rium will depend on the size of the population. For a given mutation rate, bigger pop-
ulations will have larger amounts of genetic variation at equilibrium. Population
genetic theory says that m = 2Nu, where N is the number of females and u is the rate
at which mutations occur.
We can substitute the measured values of m and u into the formula for m, to cal-
culate N. Using an estimate for the mutation rate of mtDNA of u ⬇ 0.0015 and the
average of the values for m given in Figure 13.14, we get N ⬇ 5,000. This means that
the existing amount of genetic variation within human populations today is consistent
with the drift–mutation equilibrium value for a population of 5,000 women. Data
from other genes produce both higher and lower values, and most geneticists use
10,000 as the best estimate. Obviously, this value is far lower than the size of the
human population today or even in the recent past. How can that be?
The likely answer is that a small human population expanded very rapidly some-
time in the past. When a population grows, the equilibrium amount of genetic varia-
tion also grows because genetic drift removes variation more slowly in large populations
than it does in smaller ones. If the population expands rapidly, however, it may take a
long time to reach a new equilibrium level because mutation adds variation very slowly.
Thus, the low levels of genetic variation in contemporary human populations may be
the result of a rapid expansion from a small population sometime in the past.
This conclusion is consistent with any hypothesis that posits that all contempo-
rary humans are descended from a small population living sometime in the past. A
population of 10,000 adult women corresponds to an overall population of roughly
20,000 men, women, and children. It is very difficult to see how such a small popu-
lation could have been spread across all of Africa and Eurasia. It is much more plau-
sible that the human population at the time was much larger but that modern humans
are descended from members of a single local population.
Natural selection provides an alternative explanation for the lack of genetic diver-
sity in mitochondrial genes. If a strongly beneficial mutation arose in the mitochon-
drial genome, it would rapidly sweep through the human population. Because there
is no recombination, the particular mitochondrial genotype in which the beneficial
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 363
mutation occurred would also spread, and all other mitochondrial genotypes would
disappear. In this way, natural selection could reduce the number of distinct mito-
chondrial genotypes without any reduction in the number of people. However, stud-
ies of the genes on other chromosomes are consistent with results of analyses of
mtDNA. Natural selection could produce this pattern only if a number of separate
beneficial mutations occurred throughout the nuclear genome at about the same time.
This seems implausible but cannot be ruled out as a possibility.
When human populations expanded out of Africa, they didn’t mean to colonize
Asia. They just moved next door, and next door happened to be a nearby continent.
After that population expanded, things got crowded, and another group of emigrants
moved again—not back to Africa since it was already occupied, but farther into Eura-
sia. In this way, modern human populations spread across Eurasia bit by bit. This kind
of process could have led to a very rapid geographic expansion, even if no single per-
son moved very far. For example, suppose each individual moved only 25 km in his
or her lifetime. Assuming 25-year generations and no movement backward, this would
mean that populations would spread 100 km a century, fast enough to reach Australia
from Africa in about 10,000 years.
This kind of stepwise expansion leaves a distinctive genetic signature because each
time emigrants leave their natal population, they carry a subset of the genes present
in that population. This means that the amount of genetic variation within populations
decreases as distance from Africa increases. Figure 13.15 illustrates how this works.
Initially, there are four sites, and only site 1 is inhabited. There are 32 individuals with
four different genotypes represented by the colors blue, red, purple, and green. Initially,
there are eight of each genotype. Then eight individuals from site 1 move to neigh-
boring site 2. By chance, the blue and green genotypes are overrepresented among the
emigrants. This population grows to fill the site, and because the genes are assumed
to be neutral, their frequencies are the same as among the founders. Then the process
is repeated. Eight emigrants colonize site 3, that population grows, and then site 4 is
colonized. At each step, the emigrants carry only a fraction of the genetic variation
present in their natal home. As a result, the amount of variation decreases as they move
farther and farther from their site of origin.
This is exactly the pattern of genetic variation we see in modern human pop-
ulations. Two groups of geneticists, one led by Sohini Ramachandran of Harvard
University and a second by Franck Prugnolle of the University of Cambridge, inde-
pendently computed the genetic variation in 51 populations from around the globe
using data on a large sample of microsatellite loci. Microsatellite loci are noncoding,
repetitive, and highly variable DNA sequences, often used in criminal investigations
for individual identification. They also computed the distance of each population from
East Africa along the most likely migration route from Africa. As you can see in Fig-
ure 13.16, human genetic variation decreases with distance from East Africa.
Interestingly, the same seems to be true for phenotypic variation. The Cambridge
group collaborated with Tsunehiko Hanihara of the Saga Medical School in Japan
who had compiled 37 different measurements on each of 4,666 human skulls drawn
from 105 populations around the world. For each measurement, they calculated a
measure of phenotypic variation in each population. The amount of phenotypic vari-
ation decreases with geographical distance from Africa, but the relationship is not as
strong as the relationship between genetic variation and geographical distance from
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 364
364 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
Site 1
Migration
Site 2
Population
growth
Migration
Site 3
Population
growth
Migration
Site 4
Population
growth
Africa. This makes sense because the genes that affect variation in skull morphology
should be subject to the same sampling processes that affect the microsatellite loci.
However, because the genes that affect morphology are expressed in the phenotype,
they are subject to natural selection. Local conditions favor different skull shapes, and
this partially masks the effects of the original settlement of the world.
All the copies of any extant DNA sequence that is transmitted without
recombination can be traced back to a single copy in an individual who lived
long ago.
To make this idea a bit more concrete, let’s begin by talking about the non-
recombining part of the Y chromosomes (NRY). The NRYs are just like surnames in
the United States. Traditionally, each man transmits his surname to his sons and
daughters. The names transmitted to daughters are lost when they marry; but the sons
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 365
0.9
0.8
Genetic diversity (Hs)
0.7
0.6
0.5
0.4
This figure plots the amount of
0.3 FIGURE 13.16
genetic variation of a number of
0.2 human populations from a large sample of microsatellite loci
0 5,000 10,000 15,000 20,000 25,000 30,000 against distance from East Africa along the likely path of
Geographic distance from East Africa (km) human expansion through the world. The fact that the
American populations Middle East populations
amount of genetic variation declines supports the hypothe-
Asian populations Oceanian populations sis that humans emerged from Africa and spread along the
European populations African populations hypothesized path.
carry the name, as do their grandsons, and so on. Because there is no recombination,
it’s the same with Y chromosome. Fathers transmit their Y to their sons, who trans-
mit it to their grandsons, and so on.
Now suppose that in each generation some men leave no male descendants; their
surnames are lost. Once again, the same goes for Y chromosomes. As the generations
go on, more and more names and Y chromosomes are lost until eventually every man
carries the name and the Y chromosome of a single man—the most recent common
ancestor for the Y chromosome. (Box 13.1 gives a more detailed version of this logic.)
The same process applies to the mitochondria and to any segment of DNA carried on
ordinary chromosomes that is short enough to avoid any recombination—all are
copies of the DNA segment carried by a single individual sometime in the past.
It is important to understand that each bit of nonrecombining genetic material that
we carry has its own ancestry. The most recent common ancestor of our mitochondria
is sometimes given the colorful but misleading name “Eve.” But she is not the mother
of us all, only the mother of our mitochondria. “Adam,” the most recent common
ancestor of our NRY, was not Eve’s mate; in fact, he lived tens of thousands of years
after Eve. The most recent common ancestors of each the other genes are different peo-
ple, who, as you will soon see, lived at very different times as well.
Even though the Y chromosomes of all living people are replicas of a single Y
chromosome, they are not identical because mutation occasionally introduces new
genetic variants. Since there is no recombination, once a mutation occurs in a partic-
ular man, it is carried by all of his descendants. The same goes for the mitochondrial
DNA and nonrecombining bits of autosomal genes. Mutation has two important con-
sequences. First, we can use the methods of phylogenetic reconstruction discussed in
Chapter 4 to reconstruct a tree of descent based on derived similarities. To make such
reconstructions (sometimes called gene trees) feasible, mutations must occur frequently
enough to allow the branches of the gene tree to be distinguished. Mutation also makes
it possible to estimate the time that has passed since the last common ancestor of that
gene lived, using genetic-distance measures (also discussed in Chapter 4).
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 366
Box 13.1
Mitochondrial Eve
To see why everyone carries the mitochondrion of a single To indicate that daughters inherit their mother’s mito-
woman, consider Figure 13.17. Each colored circle repre- chondria, mothers and their daughters are assigned the
sents one woman, and each row of circles represents a pop- same color. Thus, in any generation, all of the women with
ulation of 12 women during a given generation. The circles the same color carry copies of the mitochondrion of the
in row 1 represent the women in generation 1, the circles in same woman in generation 1. (This color scheme does not
the second row are women in generation 2—all of whom imply that the mitochondria carried by different women
are daughters of women in generation 1—and so on. (We have different mtDNA sequences; they could all be geneti-
ignore men because their mitochondria are not transmitted cally identical or all unique.)
to their children.) To keep things simple, we assume that In generation 1, there are three women (black, dark
population size is constant, so women must produce on gray, and light gray) who have no daughters. As a result,
average just one daughter. However, the number of daugh- after one generation, only 9 of the original 12 women and
ters born to any particular woman varies; here, we will their mitochondria have descendants. However, three of the
assume that some have no daughters, some have one, and mitochondria in the first generation have two descendants
some have two. For the moment, we will also assume that in generation 2. During the second generation, three women
there is no mutation. have no daughters. However, because one of the women
Generation “Eve”
This 8
FIGURE 13.17
mtDNA
tree for a hypothetical population of 9
12 women shows why all the mito-
chondria in any population are
10
descended from a single woman’s
mitochondria sometime in the past.
11
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 367
without a daughter has a sister who produced a daughter, one of the original women will disappear. At this point, all
copies of that woman’s mitochondria are not lost. This members of the population will be descendants of one of the
means that, by generation 3, the number of women’s mito- original females. We have run this argument forward in
chondria present in the population is reduced from 12 to 7. time, but it can also be run backward. That is to say, we can
The final outcome should be evident: as long as there are pick any generation and then move backward through time
copies of mitochondria from more than one of the original to find the single mitochondrial ancestor of every individual
women present, there is a chance that one of them will be in that generation.
lost because all the descendants of one of the women will In this example, there are only 12 women in the popu-
have no daughters. Eventually, the mitochondria of all but lation, but the same thing will happen in all populations.
Geneticists have constructed gene trees for the mitochondria, the nonrecombin-
ing portion of the Y chromosome, and a number of genes carried on other chromo-
somes. Figure 13.18 shows the gene tree based on the entire DNA sequences of the
mitochondria of several hundred people. Each branch represents a single individual.
The length of the branches are proportional to the number of mutations accumulated
along that branch. This means that the mtDNA of two individuals linked by short
branches has a more recent common ancestor than the mtDNA of two individuals sep-
arated by long branches. A node that links branches represents the common ancestor
of those individuals. The deepest and oldest nodes are labeled according to a conven-
tion adopted by geneticists and will be useful to refer to when we interpret the tree.
This tree supports the idea that modern humans originated in Africa and then
migrated out of Africa and spread across the rest of the world. Notice that all
Africans in the sample are descended from nodes L1, L2, and L3, and are connected
by the longest branches in the tree. What is not evident from this tree is that although
direct descendants of L1 are found throughout Africa, they are relatively rare; more
than two-thirds of modern Africans carry mtDNA descended from L2 and L3. All non-
Africans are descendants of two nodes, M and N, that are connected to L3 by very
short branches. The branches linking non-Africans are much shorter than branches
linking many Africans. This suggests that the population containing the last common
ancestor, L1, came from Africa. The genetic distance along branches indicates that this
individual lived about 130 kya. Later, perhap 90 to 60 kya, L2 and L3 arose in Africa
and spread thoughout the continent. Then, a population containing M and N left
Africa around 60 kya and spread across the rest of the world over the next 50 ky (Fig-
ure 13.19).
Genetic data hint that people spread first along the southern coast of Asia and
only later moved north into the rest of Eurasia.
The vast dryness of the Sahara Desert separates central and southern Africa from
Eurasia, especially 60 kya when the world was much drier than it is now. There are
only two ways around this barrier. People could have traveled up the Nile and through
present-day Palestine, or they could have moved across the southern tip of the Red
Sea and along the southern coasts of Asia. Although the southern route would not be
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 368
368 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
L1
L2
L3
150 kya: Modern humans evolve somewhere 60–80 kya: New expansion spreads L2 and L3
in Africa and L1 lineage spreads lineages. L1 reduced to a minority.
L2
L3
L1 L1
L1 L2 L3
L2 L3
L1 L1
50–60 kya: L3 gives rise to M and N lineages 30–50 kya: M and N lineages give rise to main
that spread across southern Asia geographical haplotypes in Eurasia
H JT
U
B F
A D
I
N M R N M
N N
L2 L2 U
L3 L3
L1 M M L1 M M
N N
L2 L3 L2 L3
Andaman M M Q
Islands P
N N
L2 L3 Malay L2 L3
L1 Peninsula L1
practical on foot today, sea levels were lower 60 kya, and crossing the Red Sea would
have been relatively easy. We know from the anatomically modern fossils found at
Skhūl and Qafzeh that people did manage to make their way up the Nile corridor at
least once. However, two bits of genetic evidence suggest the populations from which
all modern non-Africans are descended took the southern route. First, the amount of
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 370
370 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
genetic variation among non-Africans is hard to reconcile with more than one popu-
lation leaving Africa. This means that people either used the northern route or the
southern route, but not both. Second, some mtDNA genotypes found along the south
coast of Asia have long branches and are not found elsewhere in Eurasia. Geneticists
collected mtDNA from the Onge, an isolated hunter-gatherer population who live on
the Andaman Islands (see Figure 13.19) and from the Orang Asli, hunter-gatherers
who live in the forests of Malaysia. People in these two groups have mtDNA types that
are descended directly from the ancestral M type. This means that the last common
ancestor of the the Onge, Orang Asli, and other Asians lived 60 kya, suggesting that
both groups are descended from the earliest human colonists of Asia. Taken together,
these two facts hint that the earliest settlers of Asia came via the southern route.
It is uncertain how much gene flow there was between Asian and African
populations before 60 kya.
The gene trees constructed for mtDNA, the NRY, and a number of genes from
other chromosomes support a simple picture of the evolution of modern humans—all
people are descended from a single African population of about 10,000 people. How-
ever, there are a number of gene trees derived for other genes that are not consistent
with this picture. The most recent common ancestor of these genes lived more than
2.5 mya.
There are two different ways to explain the existence of these very old genes. The
first possibility is that there were modest amounts of gene flow from Asia or other
parts of Africa into the African population that eventually gave rise to modern
humans. This idea is supported by the fact that one gene on the X chromosome has a
tree with the deepest branches in Asia, not Africa. The other possibility is that one of
the assumptions underlying the calculations of the age of the most recent common
ancestor is wrong. If there is balancing selection, which favors heterozygous genotypes,
the methods used to reconstruct phylogenies will overestimate the age of the last com-
mon ancestor. This possibility is supported by the fact that all of the genes that pro-
duce very old dates for the last common ancestor code for proteins, which have
potentially important effects on phenotype, while all of the nuclear genes that produce
younger dates are in noncoding regions. As you will see in Chapter 14, when natural
selection favors the heterozygous genotype at a particular locus over both of the
homozygous genotypes, selection tends to preserve variation at that locus. At present,
we do not know which of these two possibilities is correct.
In the film Jurassic Park, scientists extract dinosaur DNA from the bodies of
bloodsucking insects trapped in amber and use the DNA to clone living dinosaurs. The
gap between science fiction and science has narrowed in the laboratory of Svante
Pääbo at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Ger-
many. Pääbo and his colleagues extracted and sequenced mtDNA from the original
Neanderthal fossil found in the Neander Valley of Germany in 1856. Subsequently,
the mtDNA from a number of other Neanderthal fossils has been extracted and
sequenced.
It is extraordinarily difficult to obtain mtDNA from fossils because only minute
amounts of this material remain in the fossilized bones. There are hundreds of mito-
chondria in each cell, but only two copies of each nuclear gene. This makes it even
more difficult to get nuclear DNA out of fossils. Nonetheless, in late 2006, Pääbo and
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 371
his colleagues reported sequencing more than a million bases from nuclear DNA taken
from a 38,000-year-old Neanderthal fossil found in the Vindija Cave in Croatia.
Around the same time, a Berkeley group led by Eric Rubin and Jame Noonan pub-
lished 60,000 bases from a second Neanderthal fossil. These efforts are ongoing, and
Pääbo will have a rough draft of the complete Neanderthal sequence by late 2008.
Figure 13.20 shows a gene tree based on mtDNA from seven Neanderthals and
a reference sample of modern humans from around the globe. The tree indicates that
all modern humans are equally distant from Neanderthals. This is consistent with the
idea that the modern human lineage diversified after it split from the lineage leading
to the Neanderthals.
Researchers from Pääbo’s laboratory have also computed the genetic distance
within and between Neanderthal, modern human, and ape populations. These values
indicate that the genetic distance between modern humans and Neanderthals is much
greater than the genetic distances among modern humans or among the Nean-
derthals. The magnitude of the differences between Neanderthals and modern humans
suggests that the last common mitochondrial ancestor of humans and Neanderthals
lived between 690 and 357 kya. The data also indicate that Neanderthals, like mod-
ern humans, were much less variable genetically than other apes. The relatively small
genetic distance between the three Neanderthals in Figure 13.20 indicates that their
last common ancestor lived between 352 and 151 kya.
These findings are consistent with the fossil and archaeological records of the Mid-
dle Pleistocene. Recall from Chapter 12 that the fossils from Sima de los Huesos sug-
gest that the ancestors of Neanderthals had reached Europe and begun to evolve the
distinctive Neanderthal morphology by about 300 kya. What we know about the cli-
mate and biogeography of Middle Pleistocene Eurasia and Africa suggests that these
proto-Neanderthals colonized Europe during one of the interglacial periods and were
Croatia
(Vindija)
Caucasus
Neanderthal (Mezmaiskaya)
Germany
(Neanderthal)
Africans and
non-Africans
The phylogenetic relationships of mtDNA from
FIGURE 13.20
three Neanderthals and modern humans are
plotted. The length of each horizontal branch symbolizes the genetic dis-
Modern tance between individuals or groups. The branches leading to modern
human humans represent a complex branching tree, each tip ending at a unique
Africans
mtDNA genotype of at least one modern human. This tree indicates that
the three Neanderthals are much more closely related to each other than
Africans any one of them is to any modern human.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 372
372 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
The lineages that gave rise to Neanderthals and modern humans diverged about
half a million years ago. No modern people carry mitochondria derived from Nean-
derthals living in Europe 30 kya. However, it is possible that there was some genetic
exchange between Neanderthals and humans and, therefore, that some early modern
humans carried Neanderthal mitochondria. The same calculations that indicate that
all modern human mitochondria are descended from a single individual 130 kya also
suggest that all of our mitochondria are derived from only five individuals living just
30 kya. It is possible that there was some interbreeding between Neanderthals and
modern humans: but since none of the lucky five women from whom all modern mito-
chondria are derived happened to carry mitochondria introduced by such matings, the
Neanderthal mitochondria were lost by chance. However, three lines of evidence sug-
gest that the amount of interbreeding was extremely limited.
First, Pääbo and his colleagues have attempted to extract DNA from four very
well-preserved early modern human fossils. Ironically, it is much more difficult to
obtain DNA from the fossils of early modern humans than from Neanderthals. This
is because the modern fossils are often contaminated with the DNA of all of the
archaeologists, museum curators, and molecular biologists who handle the fossil
bones. The minuscule amounts of fossil DNA may be swamped by the modern DNA,
so it’s hard to tell whose DNA has been amplified. Pääbo and his colleagues dealt with
this problem by searching for four short DNA sequences that had been found in Nean-
derthal mtDNA but not in any contemporary human samples. If these bits of Nean-
derthal mtDNA were found in the early modern humans, then there would be
evidence for interbreeding. None of the Neanderthal mtDNA as found, which suggests
that there was no interbreeding. However, since it is still possible that the early mod-
ern mtDNA samples were contaminated, this evidence is not conclusive.
Second, University of Bern geneticists Matthias Currat and Laurent Excoffier built
a detailed computer simulation of the spread of early modern humans across Europe
starting in Anatolia. In their simulation, they assume that matings could only occur
in areas where the two populations were in contact. This model indicates that even
very low rates of interbreeding would lead to high frequencies of Neanderthal mtDNA
in most of Europe. This might seem surprising because matings are restricted to the
areas in which the two populations overlap. However, the process has a cumulative
effect. To see how this works, let’s consider a simple example. Suppose the rate of inter-
breeding is 1%. After one generation of interbreeding between moderns and Nean-
derthals, the frequency of Neanderthal mtDNA in the spreading front of moderns is
1%. In the next generation, interbreeding occurs again in 1% of the matings, so the
frequency of Neanderthal mtDNA is now 2%, and so on. By the time that the front
reaches the Rhine, the frequencies of Neanderthal mtDNA are appreciable. Given the
absence of Neanderthal mtDNA in modern humans, Currat and Excoffier conclude
that there must have been a very limited amount of interbreeding over the 10,000-year
history of contact. Of course, the validity of this conclusion rests on the assumptions
in the model about what happened between 30 and 40 kya.
Finally, the availability of nuclear DNA sequence data multiplies the number of
loci about which we have data. Suppose that 30 kya, 5% of the genes in early mod-
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 373
ern populations were derived from Neanderthals. Given that all modern mitochondria
are derived from the mitochondria carried by just five individuals, it’s not so improb-
able that all the Neanderthal mtDNA were lost. However, the availability of nuclear
DNA provides us with many more genes with independent histories, and it’s much less
likely that all of them were lost by chance.
The lack of gene flow between Neanderthals and anatomically modern humans
does not necessarily mean that there was no interbreeding at all. As you will soon
learn, archaeological evidence tells us that there was enough contact between Nean-
derthals and early moderns to allow toolmaking styles to diffuse between the popu-
lations. So it seems plausible that there were ample opportunities for mating. It could
be that there were social or behavioral barriers to such matings, but it could also be
that such matings did not produce offspring.
The oldest modern human fossils have been found in Africa and date to about
190 kya.
The oldest modern human fossils were unearthed in 1963 at Omo Kibish, in south-
ern Ethiopia. One of the two skulls found there was relatively modern, while the other
was similar to Homo heidelbergensis. Originally, the dates for this site were uncertain,
but new methods now indicate a date of 190 kya. More recently, a number of fossils
were uncovered at Herto, in Ethiopia. A team led by University of California pale-
oanthropologist Tim White found the fossilized crania of two adults and one imma-
ture individual along with a number of other fragments. These skulls are intermediate
between those of modern humans and older African hominins classified as Homo hei-
delbergensis. The skulls are longer and more robust than those of most living people.
They have prominent browridges, pointed occipital bones, and other features that link
them to earlier African hominins. However, they also have some very modern fea-
tures—most notably a high, rounded braincase (Figure 13.21). Argon–argon dating
was used to date these fossils to about 160 kya.
In those days, the Herto site was on the shores of a lake, and there is good evi-
dence that hominins butchered hippopotamuses. The Middle Stone Age stone tools
found at the site are much like those found in association with earlier African Homo
heidelbergensis. Multiple unambiguous cut marks give evidence that the skulls were
defleshed by stone tools, and polished surfaces are consistent with repeated handling
by other hominins. White and his coworkers point out that the mortuary practices
used by some contemporary peoples in New Guinea leave a similar combination of
marks on skulls, and they suggest that the Herto fossils may provide early evidence of
similar types of ritual behavior.
There is evidence that modern humans also lived in other parts of Africa around
the same time, but the fossils are too fragmentary to be completely conclusive. The
most extensive finds come from the mouth of the Klasies River in South Africa. Exca-
vations at this site have yielded five lower jaws, one upper jaw, part of a forehead, and
numerous smaller skeletal fragments. Electron-spin-resonance dating indicates that the
site is 74,000 to 134,000 years old. Although one of the lower jaws clearly has a mod-
ern jutting chin and the forehead has modern-looking browridges, the fragmentary
nature of the fossils makes it difficult to be sure that these are modern humans.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 374
374 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
Modern people appeared in the Middle East and probably in Africa long before
the Neanderthals disappeared in Europe.
Even more modern fossils have been found at the Qafzeh and Skhūl caves in Israel.
Thermoluminescence and electron-spin-resonance dating techniques have shown that
these fossils are 115,000 years old. This was during a relatively warm, wet period,
when animals would have been able to move from Africa to the Middle East. This
inference is supported by the fact that other fossilized animals found at these sites are
primarily African species.
Neanderthal fossils also have been found at three other sites located quite close
to Qafzeh and Skhūl: Kebara, Tabun, and Amud. For many years there were no reli-
able absolute dates for any of these sites because they are too young for potassium–
argon dating and too old for carbon-14 dating. However, most anthropologists
assumed that the Neanderthals came first and were succeeded by the modern hominins
at Qafzeh and Skhūl. When the sites were dated by thermoluminescence and electron-
spin-resonance methods, the results were a big surprise. The Neanderthals occupied
Tabun about 110 kya, around the same time that modern humans lived nearby. How-
ever, the Neanderthals at Kebara and Amud lived 60 to 55 kya, which is about 30,000
years after anatomically modern peoples inhabited Skhūl and Qafzeh.
Additional support for the idea that Neanderthals overlapped with and were later
replaced by modern peoples comes from archaeological data. The eariest Upper Pale-
olithic tools appear at sites in the Near East between 49 and 47 kya, and spread com-
pletely across Europe by 42 to 41 kya. In most areas, it appears that anatomically
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 375
The earliest
FIGURE 13.22
Upper Pale-
olithic industry, the Aurignacian, is
found throughout Europe; three indus-
tries intermediate between the Mous-
terian and the Aurignacian (the
Châtelperronian, the Uluzzian, and the
Szeletian) are found in localized areas.
Because each of these transitional
Uluzzian Szeletian/Jerzmanovician
industries is distinctive, some scientists
0 500 km
Châtelperronian Aurignacian believe it unlikely that the Aurignacian
evolved from one of them.
modern humans and Neanderthals overlapped for only about 1,000 years. This brief
overlap suggests that the modern humans using Aurignacian tools appeared and rap-
idly replaced Neanderthals. This transition seems too rapid to be the result of genetic
change.
However, the situation is not quite so simple. In southern France, the transition
between the Mousterian and the Aurignacian is marked by the presence of a third,
intermediate industry—the Châtelperronian. Many anthropologists believe that the
Châtelperronian is the result of Neanderthals’ borrowing ideas and technology from
modern humans. University of Cambridge archaeologist Paul Mellars argues that three
kinds of evidence support this view. First, Châtelperronian tools are associated with
Neanderthal fossils at Saint-Césaire and Arcy-sur-Cure in France. Second, archaeo-
logical data indicate that the Châtelperronian and Aurignacian industries coexisted in
southern France for hundreds of years. Finally, other transitional tool kits have been
found. As Figure 13.22 shows, each of these transitional industries is localized: the
Châtelperronian in southern France, the Uluzzian in northern Italy, and the Szeletian
in central Europe. Mellars points out that each of these transitional industries is quite
distinctive, and he argues that it is not likely that the very widespread Aurignacian
evolved from several distinctive localized transitional industries.
376 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
behaved more like their European contemporaries, the Neanderthals. Then, about 60
kya there was a rapid transition to much more modern behavior in Africa, just as in
Europe. Klein believes that this behavioral transition marks the appearance of cogni-
tively modern people. About 60 kya, some key cognitive innovation evolved in at least
one human population and allowed for the rapid development of modern human
behavior—a transition commonly called the “human revolution.” Sally McBrearty of
the University of Connecticut and Alison Brooks of the George Washington Univer-
sity have challenged the idea that there was any human revolution at all. Instead, they
argue that the archaeological record from Africa shows that the complex behaviors
that appeared so abruptly in Europe 50 kya were gradually evolving in Africa between
250 and 60 kya. What seems like a revolutionary event in Europe was actually due to
the rapid replacement of Neanderthals by more technologically and culturally
advanced African immigrants.
From about 250 to 40 kya, the African archaeological record is dominated by a vari-
ety of stone tool kits that, for the most part, emphasize Mode 3 tools. These indus-
tries are collectively labeled the Middle Stone Age (MSA). Until recently, most
archaeologists saw the MSA in Africa as qualitatively similar to the much better-
known Mousterian tool industries associated with Neanderthals in Europe. Although
there is some evidence for blade production in the MSA, it was thought that the other
signatures of modern human behavior seen in Upper Paleolithic Europe were rare or
absent in Africa. According to this view, MSA peoples in Africa produced few bone
tools, developed little regional variation in tools, engaged in little long-distance trade,
didn’t bury their dead, and had no important symbolic behavior. As we will see, this
view of the MSA has been challenged by McBrearty and Brooks.
About 40 kya, new tool industries came to predominate in Africa. These indus-
tries, labeled the Later Stone Age (LSA), are distinguished by very small, carefully
shaped flakes called microliths (literally “tiny stones”) that are thought to have been
mounted on wood and bone handles to form complex, composite tools—spears, axes,
and so on. Industries of this kind are referred to as Mode 5 technology. There is also
ample evidence of ornaments, intentional burial, long-distance trade, and more inten-
sive foraging activity. The presence of Mode 5 tools, evidence for hunting large and
dangerous game, and clear proof of symbolic expression have led many archaeologists
to regard the LSA as qualitatively similar to the Upper Paleolithic in Europe.
Some evidence suggests that LSA peoples were more proficient foragers than
MSA peoples.
MSA deposits at the mouth of the Klasies River in South Africa and LSA deposits
at nearby Nelson’s Bay Cave provide a “natural experiment” that allows us to com-
pare the behavioral complexity of MSA and LSA peoples. The MSA peoples occupied
this area between 130 and 115 kya during a warm, moist interglacial period; and LSA
peoples occupied the same area about 12 kya at the beginning of the present warm,
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 377
moist interglacial period. Klein argues that several features of this comparison support
a qualitative difference between the behavior of MSA and LSA peoples.
• LSA peoples hunted larger, more dangerous game. The MSA peoples hunted large
game, particularly eland, a cow-sized antelope. They also consumed other species,
mainly buffalo and bushpig. LSA peoples also hunted large game, but they con-
centrated more on buffalo and bushpig, species that are more common but more
dangerous than eland.
• LSA peoples engaged in more sophisticated planning. MSA peoples hunted for fur
seals throughout the year, but LSA humans hunted them only seasonally. They
killed mainly young seal pups, which were available only during the short period
when the mothers and pups were on the beach. This behavior suggests that LSA
peoples could keep track of the seasons and timed their visits to the coast to coin-
cide with the availability of seal pups.
• LSA peoples engaged in more intensive foraging. The presence of artifacts that
archaeologists interpret as sinkers and fish gorges suggests that LSA peoples fished.
There is no evidence of fishing for MSA peoples.
Not all archaeologists accept these conclusions. It has been pointed out that MSA
peoples did take dangerous creatures, both at the Klasies River mouth and elsewhere.
The quantitative difference could be due to the fact that higher population densities
forced LSA peoples to take less desirable prey. There is also evidence from sites in
Sudan, the Congo, and Botswana that MSA peoples captured fish and shellfish.
McBrearty and Brooks contend that the MSA is not qualitatively similar to the
Mousterian in Europe. According to their analysis of the African archaeological
record, most of the signatures of modern human behavior were accumulated 250 kya
and 60 kya. This body of evidence has been strengthened recently by a number of new
archaeological discoveries. It seems clear that certain behavioral practices appear in
the African archaeological record thousands of years before they are seen in Europe.
Some of these developments are as follows:
• Blades appear early in the MSA. At the Kapthurin Formation in Kenya, roughly
25% of the tools are blades. Archaeologists can show that the knappers at this site
were highly skilled at blade production, wasting little raw material and making
few mistakes. This site has been dated to between 280 and 240 kya by the argon–
argon technique. Blades have also been found at a number of other MSA sites that
date between 250 and 60 kya.
• There is regional variation in MSA industries. On a continental scale, there is quite
a bit of variation within MSA industries (Figure 13.23). In North Africa, the Ater-
ian industry is characterized by tanged projectile points; the Lupemban industry
of the Congo basin is characterized by thin, long lanceolate points; and in South
Africa, the Howieson’s Poort industry is dominated by small points that fore-
shadow the LSA. We do not know whether the kind of regional variability that
characterizes the Upper Paleolithic in Europe is also characteristic of Africa
because many fewer sites in Africa have been excavated.
• Refined bone tools occur at several MSA sites. Brooks and her coworkers recov-
ered a number of exquisite bone points, some of them elaborately barbed, from
Katanda, a site on the northeastern border of the Democratic Republic of the
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 378
378 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
Pre-Aurignacian
Aterian
Nubian complex
Ethiopian MSA
Kenya Rift MSA
Lupemban
Mumba industry
Katanda MSA
Congo (Figure 13.24). The MSA layer containing the points has been dated to
between 174 and 82 kya by three independent dating methods. At Blombos Cave
in South Africa, a number of polished bone points have
been found in association with MSA tools and date to
about 72 kya.
• Composite tools were manufactured. At a site in Sudan,
which was occupied during the transition from the Early
to the Middle Stone Age, large numbers of quartz tools
have been found. Analysis of the wear patterns on these
artifacts suggest that they were hafted onto spears and
thrown or thrust into animals. The distribution of dam-
aged tools and debris at the site suggests that broken
points were brought there and replaced with new quartz
core axes.
• MSA peoples sometimes transported raw material great
distances. Most of the stone used to make tools at MSA
Beautiful bone points like this one found
FIGURE 13.24 sites comes from a short distance away. At several MSA
at Katanda in the Democratic Republic of
the Congo have been dated to between 174 and 82 kya. If these sites, however, small amounts of raw materials were
dates are correct, then Middle Stone Age peoples were capable of transported much farther. At several sites in East Africa,
producing bone tools that rival the best of the European Upper for example, tools were made from obsidian carried 140
Paleolithic tools. to 240 km (about 90 to 150 miles).
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 379
One of two
FIGURE 13.26
engraved
pieces of red ocher found at Blombos
Cave in South Africa. The artifact is
dated to 77 kya and is associated with
MSA tools. The engravings are similar
to cave art found in other parts of the
world.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 380
380 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
Images
Beads
Microliths
Mining
Barbed points
Bone tools
Fishing
Long-distance exchange
Shellfishing
McBrearty
FIGURE 13.27 Points
and Brooks
believe that many of the archaeological
Pigment processing
signatures of modern behavior in Europe
appeared gradually in Africa between Grindstones
280 and 40 kya. If they are correct,
then the genetic and cultural changes Blades
that underlie modern human behavior
accumulated gradually in Africa during
this period. kya 20 40 60 80 100 120 140 160 180 200 220 240 260 280
present-day African peoples use red ocher to decorate themselves and for symbolic
purposes, but some archaeologists have argued that red ocher might also be used
for utilitarian purposes, such as tanning hides. Few images are associated with
MSA sites, but McBrearty and Brooks suggest that this lack of artwork may be
due to the absence of deep limestone caves in which pigments are preserved.
On the basis of these data, McBrearty and Brooks concluded that there was no
human revolution in Africa. They believe that the behavioral elements that suddenly
appear with modern humans in Europe were assembled piecemeal over a period of
200,000 years in Africa (Figure 13.27). Archaeologists have been misled about what
happened because the archaeological record from Europe is much more complete than
the archaeological record from Africa. McBrearty and Brooks point out that there are
more than 10 times as many Middle Pleistocene archaeological sites in France as there
are on the entire continent of Africa. This difference is particularly striking when you
consider that Africa is about 75 times the size of France.
Others, like Klein, interpret the same evidence differently. He argues that, for the
most part, the MSA is like the Mousterian. There is evidence of a little long-distance
trade, as well as a few bone points and rare examples of shelters and art. Most MSA
sites lack these features, suggesting, he argues, that the spectacular exceptions, such
as the bone points and the ostrich shell beads, are likely to be intrusions from more
modern sites. If they really were part of the typical behaviors of MSA peoples, he
argues, they would be more common, appearing at numerous MSA sites.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 381
Box 13.2
Arcy-sur-Cure and the
Causes of Modern
Human Behavior
The Grotte du Renne, a site at the French town of Arcy-sur- recently, Shara Bailey and Hublin have analyzed the dental
Cure, provides important clues about whether the human remains from the Châtelperronian layers; these, too, are
revolution resulted from biological or cultural change in likely to have been from Neanderthals. This means that
human populations. When the site was excavated between bone tools and ornaments were used at a site occupied by
1947 and 1963, archaeologists uncovered many layers that Neanderthals. It was still possible, however, that the tools
showed evidence of human activity. The lowest—and, there- and ornaments were manufactured by anatomically modern
fore, oldest—levels yielded many Mousterian tools, and humans and subsequently acquired by Neanderthals, per-
the highest, most recent levels contained Upper Paleolithic haps by trade or by scrounging.
artifacts. In between, archaeologists found Châtelperronian This possibility was effectively eliminated by the evi-
tools and hominin fossils, including a small part of a cra- dence provided by a French team that completed a detailed
nium and a number of teeth. More significantly, they also reanalysis of the Grotte du Renne record. They discovered
unearthed a large number of worked bone tools and 36 per- that the tools and ornaments were manufactured at the site
sonal ornaments, artifacts typically associated with Upper while Neanderthals were living there. This conclusion is
Paleolithic peoples (Figure 13.28). based partly on one piece of bone that was clearly the prod-
Recently, two clever bits of detective work have shown uct of an unsuccessful effort to make an awl (a sharp tool
that the tools and ornaments at the Grotte du Renne were used to punch holes in leather), the most common bone tool
made by Neanderthals. The first piece of evidence was pro- found at the site. There were also several ornaments made
vided by a team led by Jean-Jacque Hublin of the Max from the ends of swan bones along with the matching piece
Planck Institute for Evolutionary Anthropology. They used of bone from which the ornament was cut. Taken together,
computed tomography (CT scan) to show that the shape of these data show that the tools were made by the people who
the semicircular canals of the inner ear differed between lived at the site, and these people were Neanderthals.
Neanderthals and modern humans. Using this diagnostic Thus, we are reasonably certain that both Neanderthals
feature, they were able to establish that the Grotte du Renne and anatomically modern humans carved ornaments and
cranium was from a young Neanderthal child. More tools out of bone. This conclusion affects our thinking
381
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 382
about the causes of the human revolution. The fact that the neighboring peoples. The ethnographic record tells us that
Neanderthals who lived at Arcy-sur-Cure made and used technological accomplishments do not provide a good meas-
both bone artifacts and personal adornments is not consis- ure of social or cognitive complexity. There are many cases
tent with the idea that modern behavior is caused by a cog- in the historical record of technologically more advanced
nitive innovation that Neanderthals lacked. It seems peoples coming in contact with technologically less
unlikely that Neanderthals would have been able to borrow advanced peoples. In such cases, it is common for the tech-
ideas from their neighbors if they themselves lacked sym- nologically less advanced to adopt ideas and techniques
bolic abilities. On the other hand, it is plausible that they from the more advanced without completely abandoning
might have adopted more sophisticated technology from their own way of life.
If there was a human revolution, it may have been caused by a shift in cognitive
ability that did not lead to any detectable changes in the skeleton.
Remember that the first anatomically modern humans appear in the fossil record about
100 kya. This means that if there was a human revolution, it occurred long after the
first appearance of people who look fully modern. However, we need to remember that
morphology and behavior can be decoupled. People who look fully modern could have
evolved new cognitive abilities that were not reflected in their skeletal anatomy. For
example, Klein has suggested that the human revolution may have been caused by a
mutation that allowed fully modern speech. It could be that linguistic ability evolved
late in the human lineage and gave rise to the technological sophistication and sym-
bolic behavior of the Upper Paleolithic peoples.
One problem with this explanation is that it requires very rapid genetic change.
Spoken language relies on a number of quite complicated features, such as specialized
production and processing of sound. There is considerable evidence that the brain is
specially adapted for language, and the cognitive and morphological adaptations
underlying human speech may be every bit as complicated as specialized adaptations
like as the eye. Remember from Chapter 1 that most biologists think complex, bene-
ficial adaptations are unlikely to arise by chance. Instead, complex adaptations usu-
ally require the accumulation of many small changes. It is not easy to see how such
adaptations could have arisen in just a few thousand years.
Another problem with this explanation is that it is difficult to reconcile the idea
that modern humans had unique innate abilities that gave rise to modern behavior
with recent evidence suggesting that Neanderthals adopted some of the technological
innovations of modern humans after they came into contact with them (Box 13.2).
The human revolution may have resulted from cultural, not genetic, changes. We
know that something like this happened later in human history. About 10 kya, an
equally profound transformation was associated with the adoption of agriculture.
13_Homo Sapiens [3p]:How Humans Evolved [5e] 10/6/08 4:13 PM Page 383
The gradual evolution of human behavior in Africa is more easily reconciled with
the usual picture of adaptive evolution, which is based on the gradual accumulation
of small changes. Such a process requires no special macromutations or unlikely
chance events. As we explained in Chapters 7 and 8, behavior is subject to the same
kinds of evolutionary forces that shape morphology and physiology.
Klein, a strong advocate of the human revolution, has argued that the difficulty
with this gradualist account is that it provides no ready explanation of why modern
humans waited until 60 kya to leave Africa. If, as we have seen, most of the package
of modern behavior was already assembled 100 kya, why didn’t modern humans move
north and east then, when the climate was relatively warm and moist? Instead, they
waited until Europe was cold and dry and they had to cross something like the mod-
ern Sahara to get there.
FURTHER READING
384 CHAPTER 13 • Homo sapiens and the Evolution of Modern Human Behavior
KEY TERMS
P a rt F o u r
EVOLUTION AND
MODERN HUMANS
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 386
Chapter 14
Explaining Human Variation
Variation in Traits Influenced by Single Genes
Causes of Genetic Variation within Groups
Causes of Genetic Variation among Groups
Variation in Complex Phenotypic Traits
Genetic Variation within Groups
Genetic Variation among Groups
The Race Concept
H
uman beings vary in myriad ways. In any sizable group of people, there is
variation in height, weight, hair color, eye color, food preferences, hobbies,
musical tastes, skills, interests, and so on. Some people you know are tall
enough to dunk a basketball, some have to roll up the hems of all their pants; some
have blue eyes and freckle in the sun, others have dark eyes and can get a terrific tan;
some people have perfect pitch, others can’t tell a flat from a sharp. Your friends may
include heavy drinkers and teetotalers, great cooks and people who can’t microwave
386
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 387
popcorn, skilled gardeners and some who can’t keep a geranium alive, some who play
classical music and others who prefer heavy metal.
If we look around the world, we encounter an even wider range of variation. Some
of the variation is easy to observe. Language, fashions, customs, religion, technology,
architecture, and other aspects of behavior differ among societies. People in different
parts of the world also look very different. For example, most of the people in north-
ern Europe have blond hair and pale skin, and most of the people in southern Asia have
dark hair and dark skin. As we described in Chapter 12, Arctic peoples are generally
shorter and stockier than people who live in the savannas of East Africa. Groups also
differ in ways that cannot be detected so readily. For instance, the peoples of the world
vary in blood type and the incidence of many genetically transmitted diseases. Figure
14.1 shows that the distributions of three debilitating diseases—PKU (phenylke-
tonuria, introduced in Chapter 3), cystic fibrosis, and Tay-Sachs disease—vary con-
siderably among populations. Tay-Sachs, for example, is nearly 10 times more
common among Ashkenazic Jews in New York than among other New Yorkers.
In this chapter, we will consider how much of this variation is due to genetic dif-
ferences among people. We want to know how people vary genetically within and
England
Austria
European Australia
descent Canada
New England
Black American PKU
Asian
(a)
Israel
England
Australia
New England
Ohio
African Cystic fibrosis
Asian
(b)
Ashkenazic Israel
Jewish New York
Other Israel
Jews New York
New York
Non-Jews Tay-Sachs
U.S.A. The distributions of three
FIGURE 14.1
(c) genetic diseases—(a) PKU,
0.0050 0.0100 0.0150 0.0200 (b) cystic fibrosis, and (c) Tay-Sachs—illustrate the
Gene frequency existence of variation among human groups.
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 388
among societies, and to understand the processes that create and sustain this variation.
We’ll begin by describing the nature of variation in traits that are influenced by single
genes with large effects. Next, we’ll consider variation in traits that are influenced by
many genes. As you will see, the methods that are used to assess variation in traits
caused by single genes and multiple genes are quite different. In both cases, we will con-
sider the processes that give rise to variation within and among populations. Finally,
we will use our understanding of human genetic diversity to explore the significance
and meaning of a concept that plays an important, albeit often negative, role in mod-
ern society: race. We will argue that a clear understanding of the nature and source of
human genetic variation demonstrates that race is not a valid scientific construct.
Scientists conventionally divide the causes of human variation into two categories.
Genetic variation refers to differences between individuals that are caused by the genes
that they inherited from their parents. Environmental variation refers to differences
between individuals caused by environmental factors (such as climate, habitat, and
competing species) on the organisms’ phenotypes. For humans, culture is an impor-
tant source of environmental variation.
A practical example—variation in body weight—will clarify this distinction.
Many environmental factors affect body weight. Some factors, such as the availabil-
ity of food, have an obvious and direct impact on body weight. The majority of peo-
ple living under siege in Sarajevo in the mid-1990s were undoubtedly leaner than they
were a decade earlier, when Sarajevo was a rich, cosmopolitan city. Other environmen-
tal effects are more subtle. For example, culture can affect body weight because it influ-
ences our ideas about what constitutes an appropriate diet and shapes our standards
of physical beauty. In the United States, many young women adopt strict diets and rig-
orous exercise regimens in order to maintain a slim figure because thinness is consid-
ered desirable. But in a number of West African societies, young women are secluded
and force-fed large meals several times a day for the express purpose of gaining weight
and becoming fat. In these societies, obesity is extremely desirable, and fat women are
thought to be very beautiful. Body weight also appears to have an important genetic
component. Recent research has shown that individuals with some genotypes are pre-
disposed to be heavier than others, even when diet and levels of activity are controlled.
Genetic and environmental causes of variation may also interact in complicated
ways. Consider, for example, two people who have inherited quite different genes
affecting body weight. One is easily sated, while the other craves food constantly. Both
individuals may be thin if they have to subsist on one cup of porridge a day, but only
the one who craves food will gain weight when Big Macs and fries are readily
available.
It is often difficult to separate the genetic and environmental causes of human vari-
ation in real situations. The problem is that both genetic transmission and shared envi-
ronments cause parents and offspring to be similar. For example, suppose we were to
measure the weights of parents and offspring in a series of families living in a range
of environments. It is likely that the weight of parents and offspring (corrected for age)
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 389
would be closely related. However, we would not know whether the asso-
ciation was an effect of genes or environment. Children might resemble
their parents because they inherited genes that affect fat metabolism or
because they learned eating habits and acquired food preferences from their
parents.
Quite different processes create and maintain genetic and environ-
mental variation among groups, and identifying the source of human dif-
ferences will help us understand why people are the way they are. Genetic
variation is governed by the processes of organic evolution: mutation, drift,
recombination, and selection. Biologists and anthropologists know a great
deal about how the various processes work to shape the living world and
how evolutionary processes explain genetic differences among contempo-
rary humans in particular cases.
Variation in
FIGURE 14.3
stature between
professional basketball players and profes-
sional jockeys is illustrated here by the
difference in height between Wilt Cham-
berlain (2.15 m, or 7 ft 1 in.) and Willy
Shoemaker (1.50 m, or 4 ft 11 in.).
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 390
We can prove that traits are controlled by genes at a single genetic locus by
showing that their patterns of inheritance conform to Mendel’s principles.
= = = = =
The pattern of specific language impairment (SLI) in the KE family tree sug-
FIGURE 14.4
gests that some cases of SLI are caused by a single dominant gene. Circles rep-
resent women, triangles represent men, and blue represents people with SLI. If SLI is caused by a
dominant gene, then because SLI is rare in the population as a whole, we know from the Hardy–
Weinberg equations that almost all people with SLI will be heterozygotes. Thus, Mendel’s principles tell
us that, on average, half of the offspring of a mating between a person with SLI and a person without it
will have SLI, and half of the offspring will have normal linguistic skills. Notice how well the family shown
in this tree fits this prediction.
prediction very well. Both of the children of the son without SLI are normal, and the
rest of the matings produced approximately equal numbers of normal and language-
impaired children.
Although the pattern in the KE family is consistent with the idea that SLI is caused
by a single dominant gene, it is possible that an environmental factor causes SLI to run
in families and that the observed pattern arose by chance. Scientists search for two
kinds of data to clinch the case. First, they collect data on more families. The larger
the number of families that fit the pattern associated with the inheritance of a single-
locus dominant gene, the more confident researchers can be that this pattern did not
occur by chance. Second, researchers search for genetic markers (genes whose location
in the genome is known) that show the same pattern of inheritance. Thus, if every indi-
vidual who has SLI also has a specific marker on a particular chromosome, we can be
confident that the gene that causes SLI lies close to that genetic marker.
In 1998, the Wellcome Trust researchers demonstrated that SLI in the KE family
is closely linked to a genetic marker on chromosome 7, and so it seems likely that SLI
in this family is controlled by a gene closely linked to this marker. Even more remark-
ably, the subsequent discovery of an unrelated person with the same symptoms
allowed the researchers to identify the specific gene that causes the disorder in the KE
family. The same allele of this gene, named FOXP2, was found in all affected mem-
bers of the KE family and not in 364 unrelated people without SLI. It differs from the
normal allele by a single nucleotide substitution.
Interestingly, molecular evidence suggests that the FOXP2 gene has undergone
strong directional selection since the divergence of humans and chimpanzees. Differ-
ent alleles of the FOXP2 gene are found in many animal species. The versions found
in mice and humans differ by three amino acid substitutions, while the version found
in chimpanzees differs from mice by only one of these substitutions. Since the last com-
mon ancestor of humans and mice lived more than 65 mya and the last common ances-
tor of humans and chimps only 6 mya, this suggests that there has been rapid change
in the FOXP2 gene during human evolution. Moreover, studies of variation in introns
of the FOXP2 gene and surrounding noncoding sequences suggest that the normal ver-
sion of the gene spread throughout human populations less than 200 kya, about the
same time as the origin of Homo sapiens.
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 392
The fact that SLI is caused by a single gene does not mean that the gene is respon-
sible for all of the psychological machinery in the human brain that gives rise to lan-
guage. It means only that damage to the FOXP2 gene prevents the normal
development of some of the psychological machinery necessary for language. If you
cut the wire connecting the hard disk to the power supply in your computer, the hard
disk will stop working, but that does not mean the wire contains all of the machinery
necessary for operation of the hard disk. This argument is supported by the fact that
the FOXP2 gene codes for a transcription factor belonging to a family of genes that
play an important role in regulating gene expression during development. FOXP2
itself is strongly expressed in the brains of developing fetuses. By the same reasoning,
SLI in other families may be caused by other genes whose expression is necessary for
normal brain development. Just as there are many ways to wreck your hard disk, there
may be many mutants at many loci that damage the parts of the brain necessary for
language.
Many diseases are caused by recessive genes. For example, only people who are
homozygous for hemoglobin S are afflicted with sickle-cell anemia. Other diseases
caused by recessive alleles include PKU, Tay-Sachs disease, and cystic fibrosis. All of
these diseases are caused by mutant genes that code for proteins that do not serve their
normal function, and all produce severe debilitation and sometimes death. Why
haven’t such deleterious genes been eliminated by natural selection?
One answer to this question is that natural selection steadily removes such genes,
but they are constantly being reintroduced by mutation. Very low rates of mutation
can maintain these deleterious genes because they are recessive traits, so most indi-
viduals who carry the gene are heterozygous for the deleterious genes and do not suf-
fer the disastrous consequences that homozygotes suffer. The observed frequency of
many deleterious recessive genes is about 1 in 1,000. According to the Hardy–
Weinberg equations, then, the frequency of newborns homozygous for the recessive
allele will be 0.001 ✕ 0.001 = 0.000001! Thus, only 1 in 1 million babies will carry
the disease. This means that even if the disease is fatal, selection will remove only two
copies of the deleterious gene for every 1 million people born. Because mutation rates
for such deleterious genes are estimated to be a few mutations per million gametes pro-
duced, mutation will introduce enough new mutants to maintain a constant frequency
of the gene. When this is true, we say that there is selection–mutation balance.
Some lethal genes are too common to be the result of selection–mutation balance.
In West African populations, for example, the frequency of the hemoglobin S allele is
typically about 1 in 10. How can we account for this? The answer in the case of hemo-
globin S is that this allele increases the fitness of heterozygotes. It turns out that indi-
viduals who carry one copy of the sickling allele, S, and one copy of the normal allele,
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 393
A, are partially protected against the most dangerous form of malaria, called falci-
parum malaria (Figure 14.5). As a consequence, where falciparum malaria is preva-
lent, heterozygous AS newborns are about 15% more likely to reach adulthood than
AA infants.
When heterozygotes have a higher fitness than either homozygote has, natural
selection will maintain a balanced polymorphism, a steady state in which both alleles
persist in the population. To see why balanced polymorphisms exist, first consider
what happens when the S allele is introduced into a population and is very rare. Sup-
pose that its frequency is 0.001. The frequency of SS individuals will be 0.001 ✕ 0.001,
or about 1 in 1 million, and the frequency of AS individuals will be 2 ✕ 0.001 ✕ 0.999,
or about 2 in 1,000. This means that, for every individual who suffers the debilitat-
ing effects of sickle-cell anemia, there will be about 2,000 heterozygotes who are par-
tially immune to malaria. Thus, when the S allele is rare, most S alleles will occur in (a)
heterozygotes, and the S allele will increase in frequency. However, this trend will not
lead to the elimination of the A allele. To see why, let’s consider what happens when
the S allele is common and the A allele is rare. Now almost all of the A alleles will
occur in AS heterozygotes and be partially resistant to malaria, but almost all of the
S alleles will occur in SS homozygotes and suffer debilitating anemia. The A allele has
higher fitness than the S allele when the S allele is common (Box 14.1). The balance
between these two processes depends on the fitness advantage of the heterozygotes and
the disadvantage of the homozygotes. In this case, the equilibrium frequency for the
hemoglobin S allele is about 0.1, approximately the frequency actually observed in
West Africa.
Scientists suspect that the relatively high frequencies of genes that cause a num-
ber of other genetic diseases may also be the result of heterozygote advantage. For
example, the gene that causes Tay-Sachs disease has a frequency as high as 0.05 in
(b)
some eastern European Jewish populations. Children who are homozygous for this
gene seem normal for about the first six months of life. Then, over the next few years (a) Sufferers
FIGURE 14.5
a gradual deterioration takes place, leading to blindness, convulsions, and finally of sickle-cell
death, usually by age four. There is some evidence that individuals who are het- anemia have abnormal red blood cells
erozygous for the Tay-Sachs allele are partially resistant to tuberculosis. Jared Dia- with a sickle shape. (b) Normal red
mond of the University of California, Los Angeles, points out that tuberculosis was blood cells are round. Sickle-cell anemia
partially protects against falciparum
much more prevalent in cities than in rural areas of Europe over the last 400 years.
malaria.
Confined to the crowded urban ghettos of eastern Europe, Jews may have benefited
more from increased resistance to tuberculosis than did other Europeans, most of
whom lived in rural settings.
Variation may exist because environments have recently changed and genes that
were previously beneficial have not yet been eliminated.
Some genetic diseases may be common because the symptoms they create have not
always been deleterious. One form of diabetes, non-insulin-dependent diabetes
(NIDD), may be an example of such a disease. Insulin is a protein that controls the
uptake of blood sugar by cells. In NIDD sufferers, blood-sugar levels rise above nor-
mal levels because the cells of the body do not respond properly to insulin in the blood.
High blood-sugar levels cause a number of problems, including heart disease, kidney
damage, and impaired vision. NIDD is also known to have a genetic basis. (The other
form of diabetes, insulin-dependent diabetes, occurs because the insulin-producing
cells in the pancreas have been destroyed by the body’s own immune system. It is
unlikely that insulin-dependent diabetes was ever adaptive.)
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 394
Box 14.1
Calculating Gene
Frequencies for a
Balanced Polymorphism
It is easy to calculate the frequency of hemoglobin S when common so that p is close to 1, the average fitness of the S
selection has reached a stable, balanced polymorphism. allele is close to 0; but when S is rare, its average fitness is
Suppose the fitness of AA homozygotes is 1.0, the fitness of almost 1.15. If one gene has a higher fitness than the other,
AS heterozygotes is 1.15, and the fitness of SS homozygotes natural selection will increase the frequency of that gene.
is 0, and let p be the equilibrium frequency of the allele S. If Thus, a steady state will occur when the average fitnesses of
individuals mate at random, a fraction p of the S alleles will the two alleles are equal—that is, when
unite with another S allele to form an SS heterozygote, and
a fraction 1 – p will unite with an A allele to form an AS het- 1.15(1 – p) = 1.15p + (1 – p)
erozygote. Thus, the average fitness of the S allele will be
If you solve for p, you will find that
0p + 1.15(1 – p)
1.15 – 1.10
p=
By the same reasoning, the average fitness of the A allele will 1.15 + 1.15 – 1.0
be
0.15
= ⬇ 0.1
1.30
1.15p + 1(1 – p)
which is about the observed frequency of the sickle-cell
The relationship between the average fitness of each
allele in West Africa.
allele and the frequency of hemoglobin S, shown in Figure
14.6, confirms the reasoning given in the text. When S is
1.15
Average fitness
of hemoglobin A
1
Average fitness
Average fitness
The average fitness of the S allele of hemoglobin S declines of hemoglobin S
FIGURE 14.6
as the frequency of S increases because more and more S
alleles are found in SS homozygotes. Similarly, the average fitness of the A allele of
hemoglobin A increases as the frequency of S increases because more and more A 0
alleles are found in AS heterozygotes. A balanced polymorphism occurs when the 0 p 1
average fitness of the two alleles is equal. Frequency of hemoglobin S
394
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 395
The human species inhabits a wider range of environments than any other mam-
mal. We know that natural selection in different environments may favor different
genes, and that natural selection can maintain genetic differences in the face of the
homogenizing influence of gene flow if selection is strong enough. Variation in the dis-
tribution of hemoglobin genes provides a good example of this process. Hemoglobin
S is most common in tropical Africa, around the Mediterranean Sea, and in southern
India (Figure 14.7a). Elsewhere it is almost unknown. Generally, hemoglobin S is preva-
lent where falciparum malaria is common, and hemoglobin A is prevalent where this
form of malaria is absent (Figure 14.7b). Southeast Asia represents an exception to this
pattern, and it is possible that hemoglobin E, a hemoglobin allele that is common in that
region, also provides resistance to malaria.
The digestion of lactose, a sugar found in mammalian milk, provides another
interesting example of genetic variation maintained by natural selection. Lactose is syn-
thesized in the mammary glands and occurs in large amounts only in mammalian milk.
Most mammals can digest lactose as infants but lose this ability after they are weaned.
The vast majority of humans follow the mammalian pattern because synthesis of the
necessary enzyme, lactase-phlorizin hydrolase (LHP), ceases after about age five. Such
people are said to lack lactase persistence. When people who lack lactase persistence
drink substantial amounts of fresh milk at one sitting, they suffer gastric distress that
ranges from mild discomfort to severe pain. However, most northern Europeans and
members of a number of North African and Arabian populations retain the ability to
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 396
Frequency of
hemoglobin S
0–3
3–6
6–9
9–12
12–15
>15
(a)
(b)
digest lactose as adults and are said to have lactase persistence. Evidence from family
studies indicates that the ability to digest lactose as an adult is controlled by a single
dominant gene. People who have one copy of the dominant allele are lactase-persis-
tent; those who are homozygous for the alternative recessive allele are not.
Recent molecular studies indicate that lactase persistence evolved independently
in Africa and in Europe. In 2002, Nabil Enattah and a group of Finnish researchers
showed that the mutation of a single nucleotide from C to T in a noncoding region
close to LCT, the structural gene that codes for LHP, is strongly associated with lac-
tose persistence in Finnish populations. Subsequent work showed that this mutant
allele, labeled T-13910, is associated with lactase persistence elsewhere in northern
Europe and that it regulates the expression of LCT. However, this allele is not associ-
ated with lactose persistence in Africa. In 2006, a group led by Sarah Tishkoff, now
of the University of Pennsylvania, showed why. In Africa, lactase persistence is asso-
ciated with a different mutation of a single nucleotide from G to C in the same non-
coding region. This mutant allele, labeled C-14010, also upregulates the production
of lactase in adults. Thus, the evolution of lactase persistence in Africa and Europe is
an example of a convergent adaptation.
In both Africa and Europe, high frequencies of lactase persistence only occur in
populations with a long history of dairying, suggesting that the lactase-persistence gene
evolved in response to this cultural practice. Pastoralists are people who keep livestock
and do not farm. It seems likely that in such pop-
ulations the ability to digest fresh milk is advan-
tageous. This is clearly the case among desert
pastoralists today. Gebhard Flatz of the Medi-
zinische Hochschule Hannover, Germany, studied
the Beja, a people who wander with their herds of
camels and goats in the desert lands between the
Nile and the Red Sea (Figure 14.8). During the
nine-month dry season, the Beja rely almost
entirely on the milk of their camels and goats.
They drink about 3 liters of fresh milk a day, and
they obtain virtually all of their energy, protein,
and water from milk. However, in other ecological
situations, the advantages of milk drinking are not
so obvious. There are many peoples—for example,
in southern Europe and the Middle East—who
keep cows but do not drink milk, and instead con- Pastoralists in northern Africa, such as the Beja, herd
sume cheese, yogurt, and other processed milk FIGURE 14.8
camels, and during some parts of the year they obtain vir-
products that contain little lactose. tually all of their nourishment from fresh milk. A high proportion of the Beja are
Several lines of evidence suggest that in north- able to digest lactose as adults.
ern Europe people adopted pastoralism, and then
lactase persistence evolved in response to this new ecological environment. Archaeo-
logical sites rich in the bones of domesticated cows, sheep, and goats clearly document
that pastoralism reached northern Europe about 6 kya. Two kinds of evidence suggest
that milk production played a significant role in these early pastoral economies. First,
pottery that dates to this period contains minute residues of chemical compounds
found only in milk. Second, isotopic analysis of cows’ teeth found at these early pas-
toralist sites indicates that calves were weaned from their mothers at a young age, a
practice that increases milk production. In contrast, pastoralists who emphasize meat
production allow calves to suckle longer because this increases their growth rate. These
data indicate that milk played an important role in early northern European pastoral
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 398
economies; they do not prove that lactase persistence evolved in response to milk con-
sumption. It could be that lactase persistence was already common by chance and that
these peoples adopted dairying because they could drink milk, rather than the reverse.
However, a group led by Joachim Burger of the Johannes Gutenberg University of
Mainz, Germany, was able to extract DNA from eight fossil humans from this
period, and none of these individuals carried the European lactase-persistence allele
C-13910. Therefore, it seems unlikely that lactase persistence was already common
among these early pastoralists.
Geographical patterns in the genetics of modern cattle populations also suggest that
dairying and lactose persistence coevolved. A research group led by Albano Beja-Pereira
of Université Joseph Fourier in Grenoble, France, measured the diversity of genes affect-
ing milk composition in European breeds of cattle. In his report, published 2003, Beja-
Pereira reasoned that in areas in which dairy production has been prevalent, pastoralists
would have maintained larger herds, and this would have led to greater diversity in
genes affecting milk. As can be seen in Figure 14.9, the geographical distributions of
genetic diversity affecting milk and lactase persistence correspond very closely with each
other and with the earliest archaeological site with evidence of milk production.
Until recently, the only way to determine which human genes have been subject
to selection was to guess which genetic loci might have been subject to natural selec-
(a) (b)
tion and then to test to see whether this is the case. For example, the relationship
between the sickle-cell trait and malaria was initially detected by the correlation
between the high frequency of the sickle-cell allele and the prevalence of falciparum
malaria. As of 2006, about 90 genes have been shown to be subject to recent positive
selection using this “candidate gene” approach.
The advent of complete DNA sequences has allowed scientists to detect selection
directly from these data, without any information about the function of the gene or
its prevalence in different populations. This is possible because positive selection leaves
detectable patterns in the genome. Here we focus on one of these: a new beneficial
mutation causes a selective sweep in which both the mutation and DNA linked to the
mutation on the same chromosome spread through the population. This means that
selection leading to the spread of a favorable mutation can be detected by looking at
regions of the genome in which identical long DNA sequences are common.
To understand why such sequences provide evidence for positive selection,
assume for a moment that there is no crossing over and, therefore, no recombina-
tion of genes carried on the same chromosome. Now suppose a new favorable muta-
tion arises in the population. Individuals carrying that mutation have higher fitness,
and the mutation increases in frequency. But if there were no recombination, all of
the DNA on the chromosome that happened to contain the mutant would also have
to spread, which, in turn, would result in the increase in frequency of all of the alle-
les having the good luck to be linked to the beneficial mutant. Eventually, the whole
population would carry this chromosome. Of course, this does not really happen
because crossing occurs and this shuffles the alleles at other loci. However, the rate
at which this happens depends on how closely linked the loci are—the closer they are
to the beneficial mutant the less likely it is that recombination will separate the
mutant from the allele that it was originally linked to. This means that when a new
favorable mutant initially spreads, it tends to be surrounded by a long chunk of DNA
with the same sequence. Geneticists call the sequence in some region of a chromo-
some the haplotype. Eventually, recombination breaks up this sequence, shortening
the haplotype shared by the carriers of the beneficial allele, but this takes about
10,000 years. So genes subject to recent selection are surrounded by a long haplo-
type. Alternative alleles not subject to selection are not. For example, the European
version of the lactase-persistence allele lies in the middle of a haplotype that is about
1 million base pairs long. Thus, by searching DNA sequences for long haplotypes
that are common in the population, geneticists can find sequences that recently have
been subject to strong natural selection. (See Figure 14.10 for a simple example of
this process.)
Benjamin Voight and his colleagues at the University of Chicago used this tech-
nique to scan three different human populations for signs of recent natural selection.
They used DNA sequence data on 209 individuals, 89 from Tokyo and Beijing, 60
Yoruba-speaking individuals from Ibadan, Nigeria, and 60 people of northern and
western European origin. These data, collected by a large international consortium
called the International HapMap Project, include about 800,000 polymorphic SNPs
(an SNP—single nucleotide polymorphism—is a location in the DNA sequence where
people differ by a single base.) In each population for each of these SNPs, they calcu-
lated the ratio of the length of the haplotype containing the more common allele and
the length of the haplotype containing the less common allele. Large values of the
index indicate a recent selective sweep. Using this approach, Voight and his colleagues
identified 579 regions in which a sweep is very likely to have occurred. Three-quar-
ters of these sweeps occurred in only one of the three populations. This makes sense
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 400
400
Favorable Natural
mutation selection Recombination
because most of the changes detected by this method occurred in the last 10,000 years,
long after the populations of Asia, Africa, and Europe split. The coding genes in these
regions fall into several categories:
• Reproductive system: This includes the creation of gametes and fertilization,
including genes affecting the protein structure of sperm, sperm motility, gamete
viability, and the female immune response to sperm. Genes in this category also
show rapid evolution during the divergence of humans and chimpanzees, and may
reflect ongoing male–female competition or selection for disease resistance.
• Morphlogy: Genes affecting skin color show evidence of strong selection among
Europeans, and genes affecting bone development also show rapid evolution. This
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 401
0.8
s = 0.07
s = 0.05
0.6
s = 0.03
0.4
the area of what is now Salt Lake City in 1847, but members of the church continued
to arrive until 1890. Virtually all of the immigrants were of northern European
descent. At the turn of the twentieth century, there were about 250,000 people in this
area, and about 70% of them belonged to the Mormon church. In contrast, the two
Anabaptist groups were much smaller. The founding population of the Old Order
Amish was only 200 people, and gene flow from outside the group has been very lim-
ited. Contemporary Hutterites are all descended from a population of only 443 peo-
ple and, like the Amish, have been almost closed to immigration.
Researchers studying the genetic composition of each of these populations have
found that Mormons are genetically similar to other European populations. Thus, even
though Mormon populations have been partly isolated from other European popula-
tions for over 150 years, genetic drift has led to very little change. This is just what
we would expect, given the size of the Mormon population. In contrast, the two
Anabaptist populations are quite distinct from other European populations. Because
their founding populations were small and the communities were genetically isolated,
drift has created substantial genetic changes in the same period of time.
Genetic drift can also explain why certain genetic diseases are common in some
populations but not in others. For example, Afrikaners in what is now the Republic
of South Africa are the descendants of Dutch immigrants who arrived in the seven-
teenth century. By chance, this small group of early immigrants carried a number of
rare genetic diseases, and these genes occurred at much higher frequency among mem-
bers of the colonizing population than in the Dutch populations from which the immi-
grants were originally drawn. The Afrikaner population grew very rapidly and
preserved these initially high frequencies, causing these genes to occur in higher fre-
quencies among modern Afrikaners than in other populations. For example, sufferers
of the genetic disease porphyria variegata develop a severe reaction to certain anes-
thetics. About 30,000 Afrikaners now carry the dominant gene that causes this disease,
and every one of these people is descended from a single couple who arrived from Hol-
land in the 1680s.
Overall patterns of genetic variation mainly reflect the history of migration and
population growth in the human species.
Much of the genetic variation among human groups reflects the history of the peo-
ples of the Earth. In Chapter 13 we explained that the geographical patterns of genetic
variation indicate that the human species underwent a population expansion first in
Africa about 130 kya and then throughout the world about 6 kya. There have been
several subsequent population expansions. The invention of agriculture led to expan-
sions of farming peoples from the Middle East into Europe, from Southeast Asia into
Oceania (the Pacific island groups of Polynesia, Melanesia, and Micronesia), and from
west-central Africa to most of the rest of the continent between 4,000 and 1,000 years
ago; the domestication of the horse and associated military innovations led to several
expansions of peoples living in the steppes of central Asia between 3,000 and 500 years
ago; and improvements in ships, navigation, and military organization led to the
expansion of European populations during the last 500 years.
L. L. Cavalli-Sforza, a human geneticist at Stanford University, has argued that
worldwide patterns of genetic variation preserve a record of these expansions. As
human populations expand, local populations become genetically isolated from one
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 403
another and begin to accumulate genetic differences. Of course, such local populations
usually experience gene flow with each other and with other human populations they
encounter during their expansion, and such gene flow tends to blur the effects of the
expansion. However, if there is not too much gene flow, the present patterns of genetic
variation will reflect the pattern of past migrations.
This hypothesis is consistent with a variety of different kinds of genetic data. For
example, we have already seen how patterns of variation in Y chromosomes and
mtDNA haplotypes can be used to infer the human expansions out of Africa. Espe-
cially strong support comes from recent work by two research groups, one at Cam-
bridge University and one at Stanford. Both groups worked with a sample of about
1,000 people drawn from 51 populations around the globe that were collected as part
of the Human Genome Diversity Project. For each individual, the scientists had genetic
data on a large number of microsatellite loci (377 in the Cambridge study and an addi-
tional 406 in the Stanford study). Both teams also calculated the distance along the
likely path of human expansion 60 kya from East Africa to the current location of each
population in the sample (see Figure 14.13). If contemporary patterns of human vari-
ation are due to the initial expansion of humans out of Africa, genetic distance should
be correlated with these distances, and that is exactly what the data show. Geo-
graphical distance is a very good predictor of genetic distance between human popu-
lations (Figure 14.14).
0.9
0.8
0.7
Genetic distance
0.6
0.5
0.4
0.3
0.2
0.1
0
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000
Geographic distance (km)
The genetic distance between each pair of populations plotted against the
FIGURE 14.14
distance between the populations along the likely pathway of expansion of
humans out of Africa. The two distances are strongly correlated, suggesting that the original expansion
out of Africa accounts for much of the genetic variation among different human populations. (From
Figure 2, in Andrea Manica, Franck Prugnolle, and François Balloux, 2005, “Geography Is a Better
Determinant of Human Genetic Differentiation Than Ethnicity,” Human Genetics 118:366–371.)
0.18
0.16
0.14
Proportion of men
The
0.12
FIGURE 14.15
heights
0.10 of men joining the British Army in
1939 varied considerably, illus-
0.08
trating the range of variation in
0.06 morphological characters within
0.04 populations. The tallest men join-
ing the army were more than 2 m
0.02
(around 84 in., or 7 ft) tall, and
0 50 55 60 65 70 75 80 85 others were less than 1.5 m (60
Height (in.) in., or 5 ft) tall.
In Part One, we saw that the transmission of genes from parents to offspring causes
children and parents to be phenotypically similar. If parents who are taller than aver-
age tend to have offspring who are taller than average, and parents who are shorter
than average tend to have children who are shorter than average, you might think that
height is determined by genes. In contrast, if parents and offspring are no more simi-
lar to each other than to other individuals in the population, you might think that
genes have little effect on height. The problem with this reasoning is that nongenetic
factors may also cause parents and offspring to be similar. It is known that many envi-
ronmental factors, such as nutritional levels and the prevalence of infectious diseases,
also affect height. Human parents directly affect their offspring’s environment in many
ways: they provide food and shelter, arrange for their children to be inoculated against
diseases, and shape their children’s beliefs about nutrition. Similarity between the envi-
ronments of parents and their offspring is called environmental covariation and is a
serious complication in computing heritability.
Data from studies of twins can be used to separate the effects of genetic trans-
mission from environmental covariation. The technique involves comparing the sim-
ilarity between monozygotic and dizygotic twins. Monozygotic (identical) twins
begin life when the union of a sperm and an egg produces a single zygote. Then, very
early in development, this embryo divides to form two separate, genetically identical
individuals. Dizygotic (fraternal) twins begin life when two different eggs are fertilized
by two different sperm to form two independent zygotes. Dizygotic twins share
approximately one-half of their genetic material, and they are just like other pairs of
full siblings except that they were conceived at the same time. Both monozygotic and
dizygotic twins share a womb and experience the same intrauterine environment. After
they are born, most twins grow up together in the same family. Thus, if most of the
variation in stature has a genetic origin, monozygotic twins are likely to be more sim-
ilar to one another than dizygotic twins are because monozygotic twins are genetically
identical. On the other hand, if most of the variation is due to the environment, and
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 406
the similarity between parents and offspring is due to having a common family envi-
ronment, then monozygotic twins will be no more similar to one another than dizy-
gotic twins are. Population genetic theory provides a way to use comparisons of the
similarity among dizygotic and monozygotic twins to adjust estimates of heritability
for the effects of correlated environments.
Twin studies are useful in trying to estimate the relative magnitude of the effects
of genetic variation and environmental variation on phenotypic characters, but the
data may be biased in certain ways. For example, twin studies will overestimate her-
itability if the environments of monozygotic twins are more similar than the environ-
ments of dizygotic twins. There are several reasons why this may be the case. In the
uterus, some monozygotic twins are more intimately associated than dizygotic twins
are. Monozygotic twins are always the same sex. After they are born, monozygotic
twins may be treated differently by their parents, family, teachers, and friends than
dizygotic twins are treated. It is not uncommon to see monozygotic twins dressed in
identical outfits or given rhyming names, and it is inevitable that their physical simi-
larities to one another will be pointed out to them over and over (Figure 14.16).
Studies of monozygotic and dizygotic twins suggest that somewhat more than half
of the variation in height in most human populations is due to genetic similarities
Identical, or between parents and their children.
FIGURE 14.16
monozygotic,
twins are produced when an embryo
splits at an early stage and produces two
genetically identical individuals.
Genetic Variation among Groups
Just as people within groups vary in many ways, groups of people collectively vary in
certain characteristics. For example, there is a considerable amount of variation in
average height among populations. People from northwestern Europe are tall, aver-
aging about 1.75 m (5 ft 9 in.) in height. People in Italy and other parts of southern
Europe are about 12 cm (5 in.) shorter on average. African populations include very
tall peoples like the Nuer and the Maasai, and very short peoples like the !Kung.
Within the Western Hemisphere, Native Americans living on the Great Plains of North
America and in Patagonia are relatively tall, and peoples in tropical regions of both
continents are relatively short.
Some of the variation in body size among human groups appears to be adaptive.
The fact that variation within populations has a genetic component does not
mean that differences between groups are caused solely by genetic differences.
We have seen that a significant proportion of the variation in height and body size
within some populations reflects genetic variation. We also know that the variation in
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 407
height and body shape among populations seems to be adaptive. From 960
these two facts, it might seem logical to conclude that the variation in
body size among populations is genetic, and that this variation repre-
sents a response to natural selection. 920
However, this logic is wrong! All of the variation in height within
populations could be genetic and all of the variation in height among
populations could be adaptive, but this would not mean that there is
880
variation among populations in the distribution of genes that influence
height. In fact, all of the observed variation in height among groups
genetic variation among groups. Suppose Rob and his neighbor Pete
both set out to plant a new lawn. They go to the garden store together,
buy a big bag of seed, and divide the contents evenly. Pete, an avid gar- 760
dener, goes home and plants the seed with great care. He fertilizes, bal-
ances the soil acidity, and provides just the right amount of water at just
the right times. Rob, who lives next door, scatters the seed in his back- 720
yard, waters it infrequently and inadequately, and never even considers
fertilizing it. After a few months, the two gardens are very different. 10 30 50 70 90
Pete’s lawn is thick, green, and vigorous, but Rob’s lawn hardly justifies Mean annual temperature (°F)
the name (Figure 14.18). We know that the difference between the two People living in cold climates
lawns cannot be due to the genetic characteristics of the grass seed, FIGURE 14.17
have larger bodies than those
because Rob and Pete used seed from the same bag. Nonetheless, vari- living in warm climates. The vertical axis plots mean
ation in the height and greenness of the grass within each lawn might chest girth for numerous human groups, and the horizon-
be largely genetic because the seeds within each lawn experience very tal axis plots the mean yearly temperature in the regions
similar environmental conditions. All of the seeds in Pete’s lawn get reg- in which each group lives. Because chest girth is a meas-
ular water and fertilizer, while all of the seeds in Rob’s lawn are neg- ure of overall size, these data show that people living in
lected to an equal extent. Thus, if the seed company has sold genetically colder climates have larger bodies.
Rob’s lawn
Acme
Seed
The differ-
FIGURE 14.18
ences
between two separate lawns planted
from the same bag of seed must be
environmental. However, if the seed
Pete’s lawn used was genetically variable, the differ-
ences within each lawn could be genetic.
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 408
variable seed, differences between individual plants within each lawn could be due
mainly to genetic differences between the seeds themselves.
The same argument applies to variation in human stature. The fact that much of
the variation in stature among Americans has a genetic component does not mean
that stature is determined entirely by genes. It means there is genetic variation that affects
stature, and these effects are relatively large in comparison with the effects of environ-
mental differences among Americans. It does not follow that the differences in stature
between Americans and other peoples are the result of genetic differences between them.
For example, although Americans are taller on average than citizens of Japan, these dif-
ferences in height are not necessarily genetic. That would be true only if two quite dif-
ferent conditions held. First, there would have to be a difference between Americans and
Japanese in the distribution of genes affecting stature. Second, this genetic difference
would have to be large compared with the differences in culture and environment
between the two groups. The fact that there is genetic variation among Americans does
not tell us whether or not Americans are genetically different from Japanese. The rela-
tively small effect of environmental and cultural variation on height among Americans
tells us nothing about the average difference in environment between Americans and
Japanese.
The increase in stature that coincided with modernization is evidence for the
influence of environmental variation on stature.
For height, there is good reason to believe that variation among groups is at least
partly environmental. There has been a striking effect of modernization on the aver-
age heights of many peoples. For example, Figure 14.19 plots the heights of several
groups of English boys between the ages of 5 and 21 in the nineteenth and twentieth
centuries. In 1833, 19-year-old factory workers averaged about 160 cm (about 5 ft 3
in.). In 1874, laborers of the same age averaged about 167 cm (5 ft 6 in.). In 1958,
180
1958 British average
1955 Upper social classes 1878 Public school
1874 Nonlaborers
170 1874 Laborers
1833 Factory
160
Height of boys (cm)
150
140
130
120
Height
FIGURE 14.19
increases
with time in English populations, but at 110
any given time richer people are taller.
Note that English “public schools” are
100
the equivalent of American private 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
schools. Age (years)
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 409
TABLE 14.1 Japanese men who immigrated to Hawaii during the first part of the twenti-
eth century were shorter than their children who had been born and raised in
Hawaii. The immigrants were similar in height to the Japanese who remained in Japan, which indi-
cates that the immigrants were a representative sample of the Japanese populations from which
they came. The fact that the children of the immigrants were taller than their parents shows that
environmental factors play an important role in creating variation in stature. (Data from Table
12.3, in G. A. Harrison, J. M. Tanner, D. R. Pilbeam, and P. T. Baker, 1988, Human Biology: An
Introduction to Human Evolution, Variation, Growth, and Adaptability, 3rd ed. [New York:
Oxford University Press].)
Average Sample
Height (cm) Size
the average British 19 year old stood about 177 cm (5 ft 9 in.). Similar increases in
height over the last hundred years can be seen among Swedish, German, Polish, and
North American children. These changes have occurred very rapidly, probably too fast
to be due solely to natural selection.
There also have been substantial changes in height among immigrants to the
United States in the course of a few generations. During the first part of the twentieth
century, when Japan had only begun to modernize, many Japanese came to Hawaii to
work as laborers on sugar plantations. The immigrants were considerably shorter than
their descendants who had been born in Hawaii (Table 14.1). This change in height
among immigrants and their children was so rapid that it cannot be the result of
genetic change. Instead, it must be due to some environmental difference between
Japan and Hawaii in the early twentieth century. The underlying cause of this kind of
environmental effect is not completely understood. In 1870s England, poverty was
involved to some extent, because relatively wealthy public-school boys were taller than
less affluent nonlaborers, and nonlaborers were taller than poorer laborers. Observa-
tions like this have led some anthropologists to hypothesize that increases in the stan-
dard of living associated with modernization improve early childhood nutrition and
increase children’s growth rates. However, this cannot be the full explanation. Notice
that even the richest people in England 120 years ago were shorter than the average
person in England is today. Since it seems unlikely that wealthy Britons were mal-
nourished in the 1870s, other factors must have contributed to the increase in stature
during the last 120 years. Some authorities think that the control of childhood diseases
may have played an important role in these changes.
Race is part of everyday life. For better or worse, our race affects how we see the world
and how the world sees us; it affects our social relationships, our choice of marriage
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 410
partners, our educational opportunities, and our employment prospects. We may decry dis-
crimination, but we cannot deny that race plays a major role in many aspects of our lives.
Like any widely used word, race means different things to different people. How-
ever, the understanding of race held by many North Americans is based on three fun-
damentally flawed propositions:
1. The human species can be naturally divided into a small number of distinct races.
According to this view, almost every person is a member of exactly one race; the
only exceptions are the offspring of the members of different races. For exam-
ple, many people in the United States think that people belong to one of three
races: descendants of people from Europe, North Africa, and western Asia;
descendants of people from sub-Saharan Africa; and descendants of people from
eastern Asia.
2. Members of different races are different in important ways, so knowing a per-
son’s race gives you important information about what he or she is like. For
example, biomedical researchers sometimes suggest that race predicts suscepti-
bility to diseases like high blood pressure, heart disease, and infant mortality.
Less benignly, some people believe that knowing a person’s race reveals some-
thing about that person’s intelligence or character.
3. The differences between races are due to biological heritage. Members of each
race are genetically similar to each other, and genetically different from members
of other races. Most Americans view African Americans and European Ameri-
cans as members of different races because they are marked by genetically trans-
mitted characters like skin color. In contrast, the Serbs and Croats of the former
Yugoslavia are seen by most Americans as ethnic groups, rather than racial
groups, because the differences between them are cultural rather than genetic.
Although many people believe these propositions to be true, they are not consis-
tent with scientific knowledge about human variation. There are genetic differences
between groups of people living in different parts of the world, and people from nearby
populations are more similar genetically than people from distant populations. How-
ever, as we will see, these differences do not mean that the human species can be mean-
ingfully divided into a set of nonoverlapping categories called races. The common view
of race is bad biology.
People tend to be more genetically and phenotypically similar to people who live
near them than to people who live farther away. As we have seen already in this chap-
ter, this is true for many genes. For example, the sickle-cell gene is common in central
Africa and in India but is rare elsewhere; the lactose-persistence gene, which allows
adults to digest lactose, predominates in northern Europe and parts of Africa but is
uncommon in the rest of the world. We have also seen that overall genetic similarity
is strongly associated with geographic proximity. Morphological similarities also link
neighboring peoples. For example, most of the people who live near the equator have
dark skin (Figure 14.20), and most of the people who live at high latitudes are stocky.
One consequence of these similarities among people is that it is possible to group
people into geographically based categories on the basis of their genetic or phenotypic
similarities. However, such classification schemes do not support two properties
required by the common concept of race. First, there is no natural classification scheme
because genetic variation is continuous, the placement of individuals within any sin-
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 411
This map
FIGURE 14.20
shows con-
tours in skin color. Notice that there
are smooth gradients away from the
equator.
gle category is arbitrary, and classifications based on different characters lead to rad-
ically different groupings. Second, classification schemes are not very informative. The
average difference among groups of people living in different parts of the world is
much smaller than the differences among individuals within each group; as a result,
knowing an individual’s group tells you little about her genotype.
In Chapter 4, we argued that species are distinctive entities that can be unam-
biguously identified in nature. Racial classifications for humans are quite different, and
there is no natural classification scheme for categorizing us. To see what is meant by
a natural classification scheme, consider the following analogy. Suppose you are a clerk
in a hardware store. Your boss gives you the assignment of classifying the contents of
two large cabinets. The first cabinet contains power drills made by a number of dif-
ferent manufacturers; there are many drills but only one model per manufacturer. The
second cabinet holds various different kinds of screws. The power drills vary in many
ways: they have different colors, shapes, weights, and power ratings. The screws also
vary: they have different lengths, diameters, pitches, and heads. You will have little
trouble sorting the drills into piles according to manufacturer because all the drills
made by a single manufacturer are similar in all of their dimensions: they are the same
color, shape, and weight and have the same power rating. Moreover, each model is dis-
tinctly different; there are no intermediate types. This is a natural classification system.
You will have a much harder time classifying the screws. Using length will produce one
set of piles; diameter, a second set of piles; and screw pitch, a third set of piles. More-
over, even a classification based on a single characteristic like length will require arbi-
trary distinctions: Should there be three or four piles? Should the 1.5-inch screws be
put in the pile with the smallest screws or with the next bigger size? There is no nat-
ural way to classify the screws.
You might think that the solution to this problem would be to classify people the
same way we classify organisms into species categories by using genetic measures of
shared descent. However, this strategy fails because human genetic variation varies
smoothly from one place to another. Of course, samples taken from widely separate
populations are genetically distinct. For example, as shown in Figure 14.21a, samples
taken from northern Europe, east Asia, and sub-Saharan Africa are completely
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 412
nonoverlapping and can be easily classified into three distinct groups. However, if geo-
graphically intermediate populations—for example, from southern India—are
included, this becomes impossible, as shown in Figure 14.21b. With this in mind, look
again at the plot of genetic distance versus geographic distance and notice that there
are no clumps or discontinuities that could form the basis of a natural classification
system. The Cambridge group tested this statistically and found that after controlling
for genetic distance, racial groupings added little additional predictive power.
Racial classification schemes based on different sets of characters don’t result in the
same groupings for all characters. For example, a classification scheme based on the
ability to digest lactose would yield very different groupings from one based on resist-
ance to malaria. A classification based on skin color would produce a different group-
ing from one based on height. This means that folk classification schemes based on skin
color and facial morphology are not reliable predictors of overall genetic similarity. In
Brazil, for example, people are classified according to what is called cor in Portugese.
Although the literal translaton of cor is “color,” the Brazilians’ classification is based
on more than skin color; it includes the morphology of the lips and eyes, and hair type.
Cor plays a role in Brazilian society analagous to race in North America: there is sub-
stantial prejudice against those classified as “black,” and people classified as black earn
less money than other Brazilians on average. If the Brazilian folk classification is bio-
logically meaningful, cor should be a good predictor of ancestry: “whites” should be
of mainly European ancestry, and “blacks” of mainly African ancestry.
To test this idea, Flavia Parra and his colleagues at the Universidade Federal de
Minas Gerais in Brazil assembled samples from three populations. They took blood
samples from people on São Tomé, an island near the coast of Africa. This is near the
area where the ancestors of most African Brazilians lived before being captured and
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 413
transported to Brazil as slaves. Parra’s group also took blood samples from people in
Portugal, the area where the ancestors of most European Brazilians came from. It turns
out that a small number of genetic loci differ between West African and European pop-
ulations, and this pattern can be used to compute an index of African ancestry. Finally,
these researchers collected two kinds of information from contemporary Brazilians.
They assigned each of the individuals to one of three common Brazilian categories—
“black,” “white,” and “intermediate”—on the basis of their phenotypes. They also
took blood from each individual, extracted DNA, and typed each individual for
genetic loci that differentiate West African and European populations. This analysis
allowed them to compute the extent of African ancestry for each individual. The
results are quite clear: there is very little correlation between phenotype and ancestry
(Figure 14.22). People classified as “black” and people classified as “white” on the
basis of their phenotypes have a similar range of genotypes.
Racial classification schemes explain very little of the world’s human genetic
variation.
Geneticists have tried to account for the patterns of variation across the globe. In
these studies the human species was categorized first into local groups of people
belonging to the same ethnic group, linguistic group, or nationality. Then local
groups were collected into larger, geographically based categories that correspond
roughly to the usual races. Geneticists computed the amount of variation in these char-
acters within each local group, among groups within each race, and among races. They
found that there is much more genetic variation within local groups than there is
among local groups or among races themselves. Differences within local groups
account for about 85% of all the variation in the human species. To put this another
way, suppose that a malevolent extraterrestrial wiped out the entire human species
except for one local group, which it preserved in an extraterrestrial zoo. The alien
could pick any local group at random—the Efe, the Inuit, the citizens of Ames, Iowa,
or the people of Patagonia—and then wipe out the rest of the humans on the planet.
This group would still contain, on average, 85% of the genetic variation that exists
in the entire human species.
São Tomé
Portugal
The genetic
FIGURE 14.22
index of
African ancestry for contemporary
Africans living on São Tomé (an island
“Black”
just off the coast of Gabon that served
as the entry point for the West African
slave trade), contemporary Portuguese,
Brazil “Intermediate” and contemporary Brazilians classifed as
“black,” “white,” and “intermediate” on
the basis of skin color, facial morphol-
“White” ogy, and hair texture. These data indi-
cate that the Brazilian folk classification
system predicts little about overall
Genetic index of African ancestry genetic similarity.
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 414
This fact means that knowing somebody’s racial classification does not tell you
very much about that individual’s genotype. For example, Michael Bamshad and his
colleagues at the University of Utah classified the 1,064 individuals in the CEPH (Cen-
tre d’Étude du Polymorphisme Humaine) data set (the same one used by the Stanford
and Cambridge researchers) into the four groups as defined by the U.S. Office of Man-
agement and Budget: Europeans (including western Asia), Asians (including Oceania),
Africans, and Native Americans. They then calculated the probability that two indi-
viduals randomly drawn from the same group are more similar genetically than two
individuals drawn randomly from different groups. The answers vary a bit from group
to group but cluster around two-thirds. This isn’t very good considering that random
guessing would get you this result half of the time. Moreover, the CEPH data omit a
number of geographically intermediate populations such as those in northern Africa
or India that would further weaken the predictive power of the four classifications.
And this result makes use of all 377 loci in the sample; knowing somebody’s racial
group would be even less informative about phenotypic characters affected by smaller
numbers of genes.
The bottom line is that people can be classified, and such classifications are not
necessarily arbitrary, but they do not reflect any natural subdivision of the human
species into biologically distinct groups. Nor does knowing a person’s position in a
classification reveal very much about what that person is like.
This conclusion is consistent with what we have learned about human evolution. It
seems likely that anatomically modern humans are a very recently evolved species; the
evidence from both mitochrondrial DNA and fossils suggests an age of less than 200,000
years. This means that it is less likely that natural selection and genetic drift have pro-
duced larger genetic differences within the human species than in other, older species.
Recall that the genetic differences among different species of chimpanzees are much
greater than among different groups of humans, probably because chimpanzees are a con-
siderably older genus. Moreover, we know that gene flow tends to eliminate genetic dif-
ferences between groups, and there has been extensive gene flow in human history.
Races represent cultural categories that play an important, but unfortunate, role
in society.
Many people find the conclusion that races aren’t real completely implausible.
They “know” from experience that race is real. The late Martin Luther King Jr.’s
dream that “my four little children will one day live in a nation where they will not
be judged by the color of their skin, but by the content of their character” has not yet
been realized. This is because racism is real and remains a pervasive problem in our
society and many others. But it’s important to separate the reality of racism from the
scientific understanding of the concept of race.
Many view the claim that race is not a valid scientific construct as another exam-
ple of the “political correctness” that is now common in academia. This view has some
support within anthropology. For example, in the November 1994 issue of Discover
magazine, University of Colorado anthropologist Alice Brues stated,
A popular political statement now is, “There is no such thing as race.” I won-
der what people think when they hear this. They would have to suppose that
the speaker if he were dropped by parachute into downtown Nairobi, would
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 415
FURTHER READING
Bamshad, Michael, et al. 2004. Deconstructing the relationship between genetics and
race. Nature Reviews Genetics 15:598–609.
Cavalli-Sforza, L. L., P. Menozzi, and A. Piazza. 1994. The History and Geography
of Human Genes. Princeton: Princeton University Press.
Falconer, D. S., and T. F. C. Mackay. 1996. Introduction to Quantitative Genetics. 4th
ed. Essex, Eng.: Longman.
Flatz, G. 1987. Genetics of lactose digestion in humans. Advances in Human Genet-
ics 16:1–77.
14_Human Genetic [3p]:How Humans Evolved [5e] 10/6/08 4:30 PM Page 416
1. What sources of human variation are described in this chapter? Why is it impor-
tant to distinguish among them?
2. Consider the phenotype of human finger number. Most people have exactly five
fingers on each hand, but some people have fewer. What is the source of variation
in finger number?
3. How can natural selection maintain genetic variation within human populations?
4. What is the evidence that selection has generated genetic variation between
human groups?
5. Why is it hard to determine the source of variation in human phenotypes? Why
might it be easier to determine the source of variation for other animals?
6. Explain how studies of human twins allow researchers to estimate the effects of
genetic and environmental differences on phenotypic traits.
7. What kind of body size and shape is best in a humid climate? Why?
8. Suppose you are told that differences in IQ scores among white Americans have
a genetic basis. What would that tell you about the differences in average scores
between white Americans and Americans belonging to other ethnic groups?
Why?
9. Explain why race is not a biologically meaningful category of classification.
KEY TERMS
Chapter 15
Why Evolution Is Relevant
to Human Behavior
Understanding How We Think
Inbreeding Avoidance
Human Language
Evolution and Human Culture
Culture Is a Derived Trait in Humans
Culture Is an Adaptation
T
he theory of evolution is at the core of our understanding of the natural
world. By studying how natural selection, recombination, mutation, genetic
drift, and other evolutionary processes interact to produce evolutionary
change, we come to understand why organisms are the way they are. Of course, our
understanding of evolution is far from perfect, and other disciplines, most notably
chemistry and physics, contribute greatly to our understanding of life. As the great
geneticist Theodosius Dobzhansky once said, however, “Nothing in biology makes
sense except in the light of evolution.” 417
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 418
So far, the way we have applied evolutionary theory in this book is not contro-
versial. We are principally interested in the evolutionary history of our own species,
Homo sapiens, but we began by using evolutionary theory to understand the behav-
ior of our closest relatives, the nonhuman primates. Twenty years ago, when evolu-
tionary theory was new to primatology, this approach generated some controversy, but
now most primatologists are committed to evolutionary explanations of behavior. In
Part Three, we used evolutionary theory to develop models of the patterns of behav-
ior that might have characterized early hominins. Although some researchers might
debate the fine points of this analysis, there is little disagreement about the value of
adaptive reasoning in this context. Perhaps this is because the early hominins were sim-
ply “bipedal apes,” with brains the size of modern chimpanzee brains. Not many peo-
ple object to evolutionary analyses of physiological traits, such as lactose tolerance, but
they may debate the merits of particular explanations for those traits. Similarly, most
people accept evolutionary explanations about why we live so long and mature so
slowly. These traits are clearly part of human biology, and there is a broad consensus
that evolutionary theory provides an essential key for understanding them.
The consensus evaporates when we enter the the domain of contemporary human
behavior. Most social scientists acknowledge that evolution has shaped our bodies, our
minds, and our behavior to a limited extent. However, many social scientists have been
very critical of attempts to apply evolutionary theory to contemporary human behav-
ior because they think evolutionary analyses imply that behavior is genetically deter-
mined. Genetic determinism of behavior in humans seems inconsistent with the fact
that so much of our behavior is acquired through learning, and that so much of our
behavior and beliefs is strongly influenced by our culture and environment. The notion
that evolutionary explanations imply genetic determinism is based on a fundamental
misunderstanding about how the natural world works.
All phenotypic traits, including behavioral traits, reflect the interactions between
genes and the environment.
Many people have the mistaken view that genetic transmission and learning are
mutually exclusive. That is, they believe that behaviors are either genetic and, thus,
unchangeable, or learned and, thus, controlled entirely by environmental contingen-
cies. This assumption lies at the heart of the “nature–nurture question,” a debate that
has plagued the social sciences for many years.
The nature–nurture debate is based on a false dichotomy. It assumes that there is
a clear distinction between the effects of genes (nature) and the effects of the envi-
ronment (nurture). People often think that genes are like engineering drawings for a
finished machine and that individuals vary simply because their genes carry different
specifications. For example, they imagine that basketball player Yao Ming is tall
because his genes specified an adult height of 2.3 m (7 ft 6 in.), and Earl Boykins is
short because his genes specified an adult height of 1.65 m (5 ft 5 in.).
However, genes are not like blueprints that specify phenotype. Every trait results
from the interaction of a genetic program with the environment. Thus, genes are more
like recipes in the hands of a creative cook, sets of instructions for the construction of
an organism using materials available in the environment. At each step, this very com-
plex process depends on the nature of local conditions. The expression of any geno-
type always depends on the environment. A person’s adult height is shaped by the
genes they inherited from their parents, how well nourished they were in childhood,
and the nature of the diseases they were exposed to when they were growing up.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 419
Some people understand that all traits are influenced by a combination of genes
and environment, but they reject evolutionary explanations of human behavior
because they have fallen prey to a second, more subtle, misunderstanding. Namely,
they believe that natural selection cannot create adaptations unless behavioral differ-
ences between individuals are caused by genetic differences. If this were true, it would
follow that adaptive explanations of human behavior must be invalid because there
is no doubt that most of the variation in behavioral traits, such as foraging strategies,
marriage practices, and values, is not due to genetic differences but is instead the prod-
uct of learning and culture.
This belief is false, however, because natural selection shapes learning mechanisms
so that organisms adjust their behavior to local conditions in an adaptive way. Recall
from Chapter 3 that this is exactly what happens with soapberry bugs. Male soapberry
bugs in Oklahoma guard their mates when females are scarce but not when females
are abundant. Individual males vary their behavior adaptively in response to the local
sex ratio. In order for this kind of flexibility in male behavior to evolve, there had to
be small genetic differences in the male propensity to guard a mated female and small
genetic differences in how mate guarding is influenced by the local sex ratio. If such
variation exists, then natural selection can mold the responses of males so that they
are locally adaptive. In any given population, however, most of the observed behav-
ioral variation is due to the fact that individual males respond adaptively to environ-
mental cues.
Behavior in the soapberry bug is relatively simple. Human learning and decision
making are immensely more complex and flexible. We know much less about the
mechanisms that produce behavioral flexibility in humans than we do about the mech-
anisms that produce flexibility in mate guarding among soapberry bugs. Nonetheless,
such mechanisms must exist, and it is reasonable to assume that they have been shaped
by natural selection. (Note that we are not arguing that all behavioral variation in
human societies is adaptive. We know that evolution does not produce adaptation in
every case, as we discussed in Chapter 3.) The crucial point here is that evolutionary
approaches do not imply that differences in behavior among humans are the product
of genetic differences between individuals.
In the remainder of this chapter, we will consider two ways that evolutionary the-
ory can be used to understand the minds and behavior of modern humans. As you will
see, researchers from different academic disciplines have followed different approaches
in their efforts to understand how evolution has shaped human behavior. Some have
focused on how natural selection has shaped the design of the human brain. Others
have tried to understand how the human capacity for culture, and the ability to acquire
ideas, beliefs, and values from other group members, have influenced the evolution of
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 420
human behavior. In the remainder of this chapter and the next, we will illustrate how
these approaches can be used to gain insights about the behavior of contemporary
humans.
The adaptation that most clearly distinguishes humans from other primates is our large
and very complex brain. Natural selection hasn’t just made our brains big; it has
shaped our cognitive abilities in very specific ways and molded the way we think.
Psychologists once thought that people and other animals had a few general-
purpose learning mechanisms that allowed them to modify any aspect of their phe-
notypes adaptively. However, a considerable body of empirical evidence indicates that
animals are predisposed to learn some things and not others. For example, rats quickly
learn to avoid novel foods that make them ill. Moreover, rats’ food aversions are based
solely on the taste of a food that has made them sick, not the food’s size, shape, color,
or other attributes. This learning rule makes sense because rats live in a very wide range
of environments, where they frequently encounter new foods, and usually forage at
night, when it is dark. In order to determine whether a new
food is edible, they taste a small amount first and then wait
several hours. If it is poisonous, they soon become ill, and
they do not eat it again. Rats may pay attention to the taste
of foods instead of to other attributes because it is often too
dark to see what they are eating (Figure 15.1). However,
there are limits to the flexibility of this learning mechanism.
There are certain items that rats will never sample, and in this
way their diet is rigidly controlled by genes. Moreover, the
learning process is not affected equally by all environmental
contingencies. For example, rats are affected more by the
association of novel tastes with gastric distress than they are
with other possible associations.
Natural selection determines the kinds of problems that
Rats initially sample small amounts of unfamil- the brains of particular species are good at solving. To
FIGURE 15.1
iar foods, and if they become ill soon after understand the psychology of any species, we must know
eating something, they will not eat it again. what kinds of problems its members need to solve in nature.
Our brains may be designed to solve the kinds of problems that our ancestors
faced when they lived in small foraging bands.
We know that people lived in small-scale foraging societies for the vast majority
of human history (Figure 15.2); stratified societies with agriculture and high popula-
tion density have existed for only a few thousand years (Figure 15.3). John Tooby and
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 421
Leda Cosmides of the University of California, Santa Barbara argue that complex
adaptations like the brain evolve slowly, so our brains are designed for life in forag-
ing societies. They use the term environment of evolutionary adaptedness (EEA) to
refer to the social, technological, and ecological conditions under which human men-
tal abilities evolved. Tooby and Cosmides and their colleagues envision the EEA as
being much like the world of contemporary hunter-gatherers.
People living in foraging groups face certain kinds of problems that affect their fit-
ness. For example, food sharing is an essential part of life in modern foraging groups.
Although vegetable foods are typically distributed only to family members, meat is
nearly always shared more widely. Food sharing is a form of reciprocal altruism. The
big problem with reciprocal altruism is that it is costly to interact with individuals who
do not reciprocate. Thus, Cosmides and Tooby hypothesized that human cognition
should be finely tuned to detect cheaters, and they have accumulated a convincing
body of experimental data suggesting that people are very attentive to imbalances in
social exchange and violations of social contracts.
As we have seen, there is quite a bit of uncertainty about how early humans lived,
and this adds ambiguity to predictions about human psychology based on evolution-
Evolutionary
ary reasoning. Some authorities believe that early members of the genus Homo were FIGURE 15.2
psychologists
much like contemporary human foragers. That is, they lived in small bands and sub-
believe that the human mind has evolved
sisted by hunting and gathering. They controlled fire, had home bases, and shared
to solve the adaptive challenges that
food. They could talk to one another, and they shared cultural beliefs, ideas, and tra- confront food foragers because this is
ditions. Other authorities think that the lives of the earliest species of Homo were com- the subsistence strategy that humans
pletely unlike those of modern hunter-gatherers. They think that these hominins didn’t have practiced for most of our evolu-
hunt large game, share food, or have home bases. If early hominins lived like con- tionary history.
temporary foragers, then it is reasonable to think that the human brain has evolved
to solve the kinds of problems that confront modern foragers, such as detecting free-
loaders in social exchange. On the other hand, if lifeways that characterize contem-
porary foragers did not emerge until 40 kya, then there might not have been enough
time for selection to assemble specialized psychological mechanisms to manage the
challenges that foragers face, such as food sharing.
Inbreeding Avoidance
The offspring of genetically related parents have lower fitness than the
offspring of unrelated parents do.
Geneticists refer to matings between relatives as inbred matings and contrast them with
outbred matings between unrelated individuals. The offspring of inbred matings are
much more likely to be homozygous for deleterious recessive alleles than are the off-
spring of outbred matings. As a consequence, inbred offspring are less robust and have
higher mortality than the offspring of outbred matings. In Chapter 14, we discussed a
number of genetic diseases, such as PKU, Tay-Sachs disease, and cystic fibrosis, that are
caused by a recessive gene. People who are heterozygous for such deleterious recessive
alleles are completely normal, but people who are homozygous suffer severe, often fatal
consequences. Recall that such alleles occur at low frequencies in most human popu-
lations. However, there are many loci in the human genome. Thus, even if the frequency
of deleterious recessives at each locus is very small, there is a good chance that every-
body has at least one lethal recessive gene somewhere in their genome. Geneticists have
estimated that each person carries the equivalent of two to five lethal recessives. Mat-
ing with close relatives is deleterious because it greatly increases the chance that both
partners will carry a deleterious recessive allele at the same locus. Inbreeding leads to
substantial reductions in fitness (Box 15.1). This, in turn, suggests that natural selec-
tion should favor behavioral adaptations that reduce the chance of inbreeding.
Matings between close relatives are very rare among nonhuman primates.
Box 15.1
Why Inbred Matings
Are Bad News
The following simple example illustrates why inbreeding their offspring are 33% less likely to survive to adulthood
increases the chance that offspring will be homozygous for than the offspring of outbred matings. Studies in wild pop-
a deleterious recessive gene. Suppose that Cleo has exactly ulations, though less conclusive, tell the same story. Several
one lethal recessive somewhere in her genome. If she mates studies suggest that humans are affected by inbreeding in
with a nonrelative, Mark, the chance that he carries the the same way. For example, the 161 children of father–
same deleterious recessive is simply equal to the frequency daughter or brother–sister matings studied by geneticist
of heterozygotes carrying that gene in the population. (We Eva Seemanová, of Charles University in the Czech Repub-
don’t need to consider the frequency of individuals who are lic, were twice as likely to die during their first year as their
homozygous for the recessive allele, because the allele is maternal half siblings were, and they were 10 times as likely
lethal in homozygotes.) Let p be the frequency of deleterious to suffer serious congenital defects. Other evidence comes
recessives in the population. Using the Hardy–Weinberg from studies of Moroccan Jews living in Israel who do not
law, we know that the frequency of heterozygotes is consider marriages between men and their nieces to be
2p(1 – p). The frequency of lethal recessives is typically incestuous. A study of 131 children from such marriages
about 0.001. Thus, there is only about 1 chance in 500 that indicates that they suffer a 20% reduction in fitness com-
Mark and Cleo carry the same deleterious recessive. pared with a control group from the same population.
If Cleo mates with her brother, Ptolemy, the story is
quite different. Remember from Chapter 7 that r, the coef-
ficient of relatedness, gives the probability that two indi- 100
viduals will inherit the same gene through common descent.
90
For full siblings, r = 0.5. Thus, if Cleo mates with her
Fitness reduction due to inbreeding (%)
423
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 424
Adults are not sexually attracted to the people with whom they grew up.
The fact that inbreeding avoidance is very common among primates suggests that
our human ancestors probably also had psychological mechanisms preventing them
from mating with close kin. These psychological mechanisms would disappear during
human evolution only if they were selected against. However, mating with close rela-
tives is highly deleterious in humans, as it is in other primates. Thus, both theory and
data predict that modern humans will have psychological mechanisms that reduce the
chance of close inbreeding, at least in the small-scale societies in which human psy-
chology was shaped.
There is evidence that such psychological mechanisms exist. In the late nineteenth
century, the Finnish sociologist Edward Westermarck speculated that childhood
propinquity stifles desire. By this he meant that people who live in intimate associa-
tion as small children do not find each other sexually attractive as adults. A number
of lines of evidence provide support for Westermarck’s hypothesis:
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 425
• Taiwanese minor marriage. Until recently, an unusual form of marriage was wide-
spread in China. In minor marriages, children were betrothed and the prospective
bride was adopted into the family of her future husband during infancy. There, the
betrothed couple grew up together like brother and sister. According to Taiwanese
informants interviewed by Stanford University anthropologist Arthur Wolf, the
partners in minor marriages found each other sexually unexciting. Sexual disin-
terest was so great that fathers-in-law sometimes had to beat the newlyweds to
convince them to consummate their marriage. Wolf’s data indicate that minor mar-
riages produced about 30% fewer children than did other arranged marriages
(Figure 15.6a) and were much more likely to end in separation or divorce (Figure
15.6b). Infidelity was also more common in minor marriages. When moderniza-
tion reduced parental authority, many young men and women who were betrothed
in minor marriages broke their engagements and married others.
• Kibbutz age-mates. Before World War II, many Jewish immigrants to Israel organ-
ized themselves into utopian communities called kibbutzim (plural of kibbutz). In
these communities, children were raised in communal nurseries, and they lived inti-
mately with a small group of unrelated age-mates from infancy to adulthood. The
ideology of the kibbutzim did not discourage sexual experimentation or marriage
by children in such peer groups, but neither behavior occurred. Israeli sociologist
Joseph Sepher, himself a kibbutznik, collected data on 2,769 marriages in 211 kib-
butzim. Only 14 of them were between members of the same peer group, and in
all of these cases one partner joined the peer group after the age of six. From data
collected in his own kibbutz, Sepher found no instances of premarital sex among
members of the same peer group.
• Third-party attitudes toward incest. As you may have realized already, aversions
to inbreeding extend beyond our attitudes toward our own mating behavior to
include strong beliefs about appropriate mating behavior by other individuals.
We are disgusted not only by the idea of having sex with our parents or our own
7.0
0.20
6.5
0.15
6.0 0.10
5.5 0.05
5.0 0
<1 1 2 3 4 5–9 >9 0 5 10 15 20
Wife’s age at adoption Duration of marriage (years)
(a) (b)
In minor marriages, the age of the wife when she arrives in her future husband’s
FIGURE 15.6
household (age at adoption) affects both fertility and the likelihood of divorce.
(a) The fertility of women adopted at young ages is depressed. (b) The younger a woman is when she
arrives in her husband’s household, the less likely it is that the marriage will survive.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 426
children, but also by the idea of other people having sex with their children. Daniel
Fessler and Carlos Navarrete of the University of California, Los Angeles think
that these kinds of “third-party” aversions are a form of “egocentric empathy.”
Westermarck hypothesized that coresidence during childhood generates sexual
aversions to particular partners. If that is the case, then the extent of exposure to
siblings of the opposite sex during childhood might also be linked to the strength
of feelings about one’s own behavior and the strength of feelings about the behav-
ior of others. Debra Lieberman, John Tooby, and Leda Cosmides at the Univer-
sity of California, Santa Barbara, along with Fessler and Navarrete, have tested
these predictions by conducting experiments in which they asked subjects (uni-
versity undergraduates) to contemplate consensual sibling incest involving hypo-
thetical adults. The results from both studies largely confirmed Westermarck’s
hypothesis. Those who had grown up with opposite-sex siblings had stronger neg-
ative responses to the hypothetical scenario than those who had not. Moreover,
women generally had stronger aversive responses to the hypothetical scenario than
men did.
Human Language
Languages are clearly part of human culture; children learn to speak the language of
the people around them. Although people in different parts of the world speak dif-
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 427
ferent languages, they share the morphological adaptations that make spoken language
possible and the cognitive capacities that allow them to learn and to use language. The
abilities to decode words and to extract meaning from combinations of words are
based on highly specialized and derived cognitive mechanisms.
Humans can perceive speech sounds at a much more rapid rate than other
sounds.
Phonemes are the basic unit of speech perception, the smallest bits of sound that
we recognize as meaningful elements of language. Not all languages recognize the same
phonetic distinctions. For example, English speakers hear the difference between “L”
and “R” sounds, but Swahili and Japanese speakers lump them together in a single
phoneme. Interestingly, infants can initially distinguish the full range of phonemes, but
during the first year of life they become insensitive to distinctions that are not part of
the language spoken around them.
We are able to decode very rapid streams of phonemes. When we speak normally,
we utter about 10 phonemes per second. If we concentrate, we can understand peo-
ple who bombard us with 25 phonemes per second. By contrast, even the best Morse
code operators cannot keep up once the transmission rate exceeds about 5 clicks per
second. In fact, if the rate of transmission of a simple sound like a Morse code click
exceeds 20 sounds per second, we hear a continuous buzz rather than a sequence of
discrete sounds.
It’s not hard to imagine why the ability to communicate rapidly was favored by
natural selection. The early human who could decode the hurried shout “Look out, a
lion is behind you” would have a substantial edge over the poor fellow who could only
say, “Speak more slowly, I can’t . . . ”
Consider the following two sentences: Org killed your brother and Your brother
killed Org. The words are the same, but the meanings produced in your brain are very
different. One situation calls for revenge, and the other calls for keeping a close watch
on Org’s relatives. We hear speech as a sequence of words, like cars coming down an
off-ramp. In order to create meaning, our brain interprets this information using rules
that specify meaning. These rules are what we call grammar. These are not the obscure
rules that modern Western people struggle to master in school, but unconscious rules
that allow almost every person to speak and understand spoken language effortlessly.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 428
The world’s languages share a number of grammatical rules that allow people to
use sequences of words to express useful ideas. Some of these rules were listed in a
paper written by Steven Pinker, now at Harvard University, and Paul Bloom, now at
Yale University:
• Words must belong to categories such as “noun,” “adjective,” and “verb,” which
correspond to basic features of the world that people need to talk about—in this
case, things, the qualities of things, and actions, respectively.
• Some rules allow the listener to figure out what the speaker intended to say about
the relationship between things, qualities, and actions. These kinds of rules tell us
that Org is a killer in the first sentence above and a corpse in the second. In Eng-
lish, these rules are based on the order of the words: the actor comes before the
verb, and the recipient of the act comes after the verb. In other languages these
distinctions are made by modification of the nouns to indicate, for example, which
person is the killer and which is the victim. Linguists refer to this part of gram-
mar as syntax.
• Words are grouped into phrases that also belong to categories such as “noun
phrase” and “verb phrase.” By combining words in this way, language can
express a wide range of thoughts with a small number of words. The noun brother
by itself is not very informative. By combining brother with other words in noun
phrases, however, you can refer to my brother, your brother, your older brother,
your husband’s brother, and so on.
• Some rules allow a listener to determine which words the speaker meant to group
together in phrases. These rules allow us to distinguish between the sentences Your
brother killed my father and My brother killed your father by telling us that your
is to be grouped with brother in the first sentence and with father in the second
sentence.
• Verbs are modified to indicate when events occurred. For example, we can dis-
tinguish between the statements Org killed your brother and Org will kill your
brother. Verbs are also modified to indicate the relative timing of two events.
Thus, Org had killed your brother before Thag showed up tells the listener that
Thag did not witness the killing.
There are other grammatical rules for forming questions, for indicating states of mind,
and for reducing the amount of memory necessary to process a sentence.
This system is not perfect. Even with all the grammatical rules in action and with
help from context, many sentences allow for more than one meaning. Wordplay such
as double entendre relies on such ambiguity. Nonetheless, by combining discrete words
using grammatical rules, humans can communicate an incredible range of useful infor-
mation (and make very bad puns).
The psychological mechanisms that structure language are different from other
cognitive abilities.
ized adaptation; it is just another complex skill acquired by the same general learning
mechanism.
This point of view does not take into account several lines of evidence that sug-
gest that we have special-purpose cognitive mechanisms specifically shaped by natu-
ral selection to allow people to learn and to use language. First, we would expect to
find a direct relationship between cognitive abilities and linguistic skills if the ability
to learn and to use language were based solely on general cognitive ability. However,
this is not the case. Some stroke victims lose nearly all of their linguistic ability, but
their intelligence is not diminished. These people are not fluent language users, and
they have particular difficulty with syntax, but they do not show any other cognitive
deficiencies. There are also people who speak grammatically but are otherwise men-
tally impaired. For example, many children suffering from spina bifida are severely
retarded, and they cannot learn to read, write, or do simple arithmetic. Nonetheless,
some of them speak fluently and at length, using grammatically correct sentences, but
the content of their speech is nonsensical.
Second, if the ability to use language were like other complex skills, children
would need considerable tutoring to learn grammatical rules. But children seem to
know some universal grammatical principles without having to learn them. Most chil-
dren begin to acquire the grammar of their native language when they are about two
years old, and by the time they are three, most children make relatively few gram-
matical errors. Many linguists believe that children have to learn the surface gram-
mar of their own language—how to speak properly in English, Spanish, or Farsi, for
example—but they do not need to learn the deep, universal grammar that underlies
all languages. This knowledge is innate. In fact, these linguists think that without some
innate knowledge of the structure of language, children could not learn language at
all.
TABLE 15.1 Responses of vervet monkeys to playbacks of alarm calls. The data indicate
that the monkeys recognize the meaning of different calls. (From Table 1, in
R. M. Seyfarth, D. L. Cheney, and P. Marler, 1980, “Monkey Responses to Three Different Alarm
Calls: Evidence of Predator Classification and Semantic Communication” Science, 210: 801–803.)
RESPONSE TO PLAYBACK
Type of Climb into Run into
Alarm Call Tree Bushes Look Up Look Down
Leopard 8 2 4 1
Eagle 2 6 7 4
Snake 2 2 2 14
versity of California, Los Angeles emphasizes that ape abilities are comparable to those
of very young children just beginning to acquire language, and she suggests that there
is continuity between the linguistic abilities of apes and humans. Others, including
Steven Pinker, emphasize differences between apes and young children in the ease with
which they learn symbols, acquire grammatical rules, and naturally grasp the com-
plexities of syntax. Those that emphasize this gap conclude that linguistic abilities in
humans are derived traits.
In the last few years, primatologists have documented a range of behavioral variation
across groups in chimpanzees, orangutans, and capuchin monkeys. For example, chim-
panzees living on the western shores of Lake Tanganyika raise their arms and clasp
hands while they groom, but chimpanzees living on the eastern shores of the lake don’t.
Orangutans in some areas use sticks to pry seeds out of fruits, but at other sites, orang-
utans are unable to extract the seeds because they have not mastered this technique.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 432
Cebus monkeys show considerable variation in foraging techniques and social con-
ventions. For example, capuchins at some sites participate in long bouts of mutual
hand sniffing (Figure 15.12), but capuchins at other sites never display this behavior.
In some cases, scientists have documented the appearance, diffusion, and eventual
extinction of behavioral variants.
(a) (b)
Bamboo rafts like the one shown in (a) may have been the precursors of the
FIGURE 15.11
great Chinese junks (b).
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 433
Box 15.2
Understanding the
Sources of Behavioral
Variation
In the early 1960s, Jane Goodall reported that chimpanzees When researchers began to study chimpanzees at other
at Gombe stripped the leaves from slender twigs and used sites in Africa, they discovered new forms of tool use. More-
the twigs as probes to fish for termites. As Goodall contin- over, they began to realize that the behavioral repertoire of
ued her work, she documented a number of additional chimpanzees varied across groups. For example, in the
forms of tool use. For example, chimpanzees at Gombe also Mahale Mountains, just miles away from Gombe, chim-
use long, slender twigs to collect ants, crumple leaves and panzees raise their arms and clasp hands while they groom
use them as “sponges” to soak up water from tree hollows, (Figure 15.16), a position never adopted at Gombe. As more
and use leaves to wipe debris from their bodies. These chimpanzee communities were habituated and studied,
observations made a huge impact in anthropology and the more and more examples of behavioral differences were
social sciences because tool use was thought to be a defin- accumulated.
ing feature of the human species. Darwin himself thought In 1999, a group of researchers led by Andrew Whiten
that bipedal evolution had been favored by natural selection of the University of St. Andrews published an exhaustive cat-
because it enabled early hominins to carry their tools from alogue of behavioral variants observed in seven well-studied
one place to another. But the discovery that tool use occurs
in a wide range of species—from Caledonian crows that
fashion hooks out of twigs (Figure 15.14), to sea otters that
use rocks to bash open abalone shells, and capuchin mon-
keys that use large stones to crack hard-shelled nuts (Fig-
ure 15.15)—demonstrates that humans do not have a
monopoly on technology.
In northeastern Brazil,
FIGURE 15.15
Cebus monkeys use large
stones to smash open hard-shelled nuts on sand-
stone slabs. This behavior is now being studied by a
Caledonian crows make and use tools in group of researchers, including Dorothy Fragaszy,
FIGURE 15.14
nature. Here a female has bent the end of a Elisabetta Visalberghi, Patrícia Izar, and Eduardo
twig and is using the hooked end to retrieve a food reward. (Photo Ottoni of the Universidade de São Paulo, and
courtesy of the Behavioural Ecology Research Group, Department of Marino Gomes de Oliveira of Fundação BioBrasil.
Zoology, University of Oxford.) (Photo by Peter Oxford.)
435
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 436
long, slender twigs (or wands) for ant dipping, but chim-
panzees in the Ivory Coast’s Taï Forest typically choose
shorter sticks. Gombe chimpanzees hold the wand in one
hand and pull it slowly through the other hand, and then
pop the ants into the mouth. Taï chimpanzees hold the wand
in one hand, then use their lips to wipe the ants directly off
the wand.
Whiten and his colleagues note that “it is difficult to see
how such behaviour patterns could be perpetuated by social
learning processes simpler than imitation.” According to
this view, chimpanzees at Taï and Gombe use different tech-
niques because they see others in their communities using
those techniques, and they imitate the behaviors that they
have observed. Close observation also seems to play a role
in the development of termite fishing in chimpanzees. Eliz-
abeth Lonsdorf and her colleagues at the University of Min-
nesota videotaped young chimpanzees while their mothers
were fishing for termites. She found that young females
watched their mothers carefully as they fished for termites,
Chimpanzees in the but young males were considerably less attentive (Figure
FIGURE 15.16 15.17). Lonsdorf also discovered that mothers varied in
Mahale Mountains in
Tanzania often hold their hands above their heads their fishing techniques, some consistently using longer
and clasp their partner’s hands as they groom. This twigs than others. Daughters tended to use the same kinds
grooming posture has never been seen at Gombe, of tools that their mothers used, but sons did not.
just 135 km (about 90 miles) away. (Photograph You might be convinced that observational learning is
courtesy of William C. McGrew.) the only plausible explanation for differences in ant-dipping
techniques at Taï and Gombe. After all, it seems implausi-
ble that all the chimpanzees at Gombe would adopt one
wild chimpanzee communities. Their list includes 39 behav- method, while all the chimpanzees at Taï would adopt a dif-
iors that are common in some communities but absent in ferent method, if each chimpanzee figured out the technique
others. These behaviors include various forms of tool use,
distinctive grooming styles, and courtship gestures. Some
behaviors, such as fishing for algae, occur in only one com-
munity; other behaviors, such as dipping for ants, occur in
multiple communities. In some cases, behavioral variants
seem to be regionally specific. Thus, chimpanzees use stones
to hammer open hard-shelled nuts in West Africa but not in
East Africa. Other behaviors, such as rain displays that are
performed at the beginning of rainstorms, occur in widely
separated locations.
In some cases, chimpanzees in different groups perform
similar tasks in different ways. For example, chimpanzees
use different techniques to collect safari ants. Safari ants live
in subterranean nests and often migrate through the under-
brush in dense swarms. Columns of marching ants, num-
bering hundreds of thousands of individuals, provide a Young female chimpanzees carefully watch
mobile snack for chimpanzees. Safari ants aggressively FIGURE 15.17
their mothers termite fishing, and they tend
defend themselves and have a painful bite. Chimpanzees to acquire the same kinds of techniques that their mothers use. Males
apparently find ants tasty but don’t like being bitten, so they are much less attentive to their mothers and do not match their
use tools to gather them. Chimpanzees at Gombe choose mother’s techniques.
436
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 437
for ant dipping on its own. This makes sense as long as there
is no functional reason why one technique works better than
another technique. However, Tatyana Humle, now at the
University of Wisconsin, and Tetsuro Matsuzawa of Kyoto
University, have discovered that functional distinctions do
underlie differences in ant-dipping techniques in a small
community of chimpanzees in Bossou, Guinea. The Bossou
chimpanzees are accomplished tool users. They use stones to
hammer open nuts, they use sticks to scoop algae off the
water surface, and they also dip for ants.
The Bossou chimpanzees’ ant-dipping tools are inter-
mediate in length between the tools used at Gombe and
those used at Taï. Sometimes Bossou chimpanzees use the
two-handed pull-and-wipe method like the Gombe chim- A chimpanzee in Bossou uses its mouth to
panzees, and sometimes they wipe the tool with their FIGURE 15.18
collect ants from a wand.
mouths like the chimpanzees at Taï do (Figure 15.18).
Humle and Matsuzawa discovered that the chimpanzees at
Bossou eat two different kinds of ants: red ones and black like the black ants at Bossou, and the ants that the chim-
ones. The black ants are considerably more aggressive than panzees eat at Taï are like the milder red ants at Bossou.
the red ants, and their bite is more painful; both kinds of Humle and Matsuzawa’s observations are important
ants are more aggressive at their subterranean nests than because they remind us just how hard it is to figure out how
when they are migrating. The chimpanzees use longer tools behaviors are acquired and transmitted in natural settings.
to dip for black ants than to dip for red ones, and they use It is possible that behavioral differences across chimpanzee
longer tools when they dip at the entrance to the nest than communities represent cultural traditions passed from one
when they dip into the migrating swarms. They are also generation to the next as naïve youngsters observe skilled
more likely to dip from a position of safety off the ground adults and imitate their behavior. However, it is also possi-
when they dip at the nest site than when they dip on migrat- ble that behavioral differences arise as individuals within
ing ants. The chimpanzees used the pull-through technique groups encounter similar conditions, gain exposure to the
mainly when they used long tools at ant nests. They tended artifacts that others use, and then learn the skill on their
to use the mouthing technique when they employed shorter own. Many different learning mechanisms could produce
sticks to feed on migrating ants. Thus, the chimpanzees at the kinds of stable differences in behavior that researchers
Bossou apparently use safer techniques for more dangerous have documented among different chimpanzee groups, and
prey. Humle and Matsuzawa think that there is good reason it takes careful work under controlled conditions to dis-
to believe that the ants that chimpanzees eat at Gombe are criminate between them.
Culture Is an Adaptation
Chimpanzees and capuchins are among the world’s cleverest creatures. In nature, they
use tools and perform many complex behaviors; in captivity, they can be taught
extremely demanding tasks. Chimpanzees and capuchins live in social groups and have
ample opportunity to observe the behavior of other individuals, and yet the best evi-
dence suggests that neither chimpanzees nor capuchins make much use of observa-
tional learning. Thus, observational learning appears not to be simply a by-product of
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 438
The archaeological record suggests that during the Pleistocene, humans occupied
virtually all of Africa, Eurasia, and Australia. Modern hunter-gatherers developed an
Infants are astounding variety of subsistence practices and social systems. Consider just a few
FIGURE 15.19
prone to examples. The Copper Eskimo lived in the high Arctic, spending their summers hunt-
spontaneous imitation of the behaviors ing near the mouth of the Mackenzie River and the long, dark months of the winter
they observe. Here, a 13-month-old living on the sea ice and hunting seals. Groups were small and heavily dependent on
infant flosses her two teeth.
hunting. The Xo lived in the central Kalahari Desert of Botswana, collecting seeds,
tubers, and melons, hunting impala and gemsbok, enduring fierce heat, and living
without surface water for months at a time. Their small, nomadic bands were linked
together in large clusters organized along male kinship lines. The Chumash lived on
the southern California coast, gathering shellfish and seeds and fishing the Pacific from
great plank boats. They lived in large permanent villages with pronounced division of
labor and extensive social stratification.
The fact that the Xo could acquire the knowledge, tools, and skills necessary to
survive the rigors of the Kalahari is not so surprising; many other species live there,
too. What is amazing is that the same brain that allowed the Xo to survive in the Kala-
hari also permitted the Copper Eskimo to acquire the very different knowledge, tools,
and skills necessary to live on the tundra and ice north of the Arctic Circle, and the
Chumash to acquire the skills necessary to cope with life in crowded, hierarchical set-
tlements. No other animal occupies a comparable range of habitats or utilizes a com-
parable range of subsistence techniques and social structures. For example, savanna
baboons, the most widespread primate species, are limited to Africa and Arabia, and
the diet, group size, and social systems of these far-flung baboon populations are much
more similar to one another than the diet, group size, and social systems of human
hunter-gatherers are.
Humans can live in a wider range of environments than other primates because
culture allows us to accumulate better strategies for exploiting local environments
much more rapidly than genetic inheritance can produce adaptive modifications. Ani-
mals like baboons adapt to different environments using various learning mechanisms.
For example, they learn how to acquire and process the food they eat. Baboons in the
lush wetlands of the Okavango Delta of Botswana learn how to harvest roots of water
plants and how to hunt young antelope during the birth season. Baboons living in the
harsh desert of nearby Namibia must learn how to find water and process desert foods.
All such learning mechanisms require prior knowledge about the environment: where
to search for food, what strategies can be used to process the food, which flavors are
reinforcing, and so on. More detailed and more accurate knowledge allows more accu-
rate adaptation because it allows animals to avoid errors and acquire a more special-
ized adaptation.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 439
In most animals, this knowledge is stored in the genes. Imagine that you captured
a group of baboons from the Okavango Delta and moved them to the Namibian
desert. It’s a very good bet that the first few months would be tough for the baboons;
but after a relatively short time, the transplanted group of baboons would probably
be quite similar to their neighbors. They would eat the same foods, have the same
activity patterns, and use the same grooming techniques. They might even adjust the
size of their groups and the size of their social networks. The transplanted baboons
would become similar to the local baboons because they acquire a great deal of infor-
mation about how to be a baboon genetically; it is hardwired. To be sure, the trans-
planted baboons would have to learn where things are, where to sleep, which foods
are desirable, and which foods to avoid, but they would be able to do this without con-
tact with local baboons because they have the basic knowledge built in.
Human culture allows accurate adaptation to a wider range of environments
because cumulative cultural adaptation provides more accurate and more detailed
information about the local environment than genetic inheritance systems can provide.
People are smart, but individual humans can’t figure out how to live in the Arctic, the
Kalahari, or anywhere else without drawing on the knowl-
edge of local people. Think about being plunked down on an
Arctic beach with a pile of driftwood and seal skins and try-
ing to make a kayak (Figure 15.20). You already know a lot:
what a kayak looks like, roughly how big it is, and some-
thing about its construction. Nonetheless, you would almost
certainly fail. And suppose you did make a passable kayak.
You would still have dozens of other skills to master before
you could make a contribution to the Inuit economy. The
Inuit could make kayaks and do all the other things they
needed to do to stay alive in a challenging environment,
because they could make use of a vast pool of useful infor-
mation stored in the minds of other people in their popu-
lation.
The information contained in this pool is accurate and
adaptive because the combination of individual learning The people of Nunivak Island in the Bering
FIGURE 15.20
and observational learning leads to rapid, cumulative adap- Sea hunted seals from kayaks like this one.
tation. Even if most individuals blindly imitate the behavior These kayaks weighed less than 15 kg (about 30 lb) and could be
of others, some individuals may occasionally come up with paddled at speeds of up to 11 km (7 miles) per hour.
a better idea, and this will nudge traditions in an adaptive
direction. Observational learning preserves the many small nudges and exposes the
modified traditions to another round of nudging. This process generates adaptation
more quickly than genetic inheritance does. The complexity of cultural traditions can
explode to the limits of our capacity to learn them, far past our ability to make care-
ful, detailed decisions about them.
The importance of culture in human affairs has led many anthropologists to con-
clude that evolutionary thinking has little to contribute to understanding human
behavior. They argue that because evolution shapes genetically determined behaviors
but not behaviors that are learned, culture is independent of biology. This argument
is a manifestation of the nature–nurture controversy, and we explained why this rea-
soning is flawed at the beginning of the chapter. Although many anthropologists have
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 440
rejected evolutionary thinking about culture, many evolutionists have made the oppo-
site mistake. They reject the idea that culture makes any fundamental difference in the
way that evolution has shaped human behavior and psychology. Some evolutionists
argue that the genes underlying the psychological machinery that gives rise to human
behavior were shaped by natural selection, so, at least in ancestral environments, the
machinery must have led to fitness-enhancing behavior. If the adaptation doesn’t
enhance fitness in modern environments, that’s because our evolved psychology is
designed for life in a different kind of world.
We think both sides in this argument are wrong. We can’t understand humans
without understanding the complex interplay between biology and culture. This is
because cumulative cultural evolution is rooted in a novel evolutionary trade-off
between benefits and costs. Observational learning is beneficial because it allows
human populations to accumulate vast reservoirs of adaptive information over many
generations, leading to the cumulative cultural evolution of highly adaptive behaviors
and technology. Because this process is much faster than genetic evolution, it allows
human populations to evolve cultural adaptations to local environments: kayaks in the
Arctic and blowguns in the Amazon. The ability to adjust rapidly to local conditions
was highly adaptive for early humans because the Pleistocene was a time of extremely
rapid fluctuations in world climates. However, the psychological mechanisms that cre-
ate this benefit come with a built-in cost. Remember that the advantage of observa-
tional learning is that it avoids the need for all individuals to figure out everything for
themselves. They can just do what others do. But to get the benefits of social learning,
people have to be credulous, generally accepting that other people are doing things in
a sensible and proper way.
This credulity opens up human minds to the spread of maladaptive beliefs and
behaviors. If our neighbors believe that it’s beneficial to bleed sick people and that it’s
a good idea to treat corn with lye, we believe that, too. This is how we get wondrous
adaptations like kayaks and blowguns. But we have little protection against the per-
petuation of maladaptions that somehow arise. Even though the capacities that give rise
to culture and shape its content must be (or at least must have been) adaptive on aver-
age, the behavior observed in any particular society at any particular time may reflect
evolved maladaptations. Examples of these sorts of maladaptations are not hard to find.
Maladaptive beliefs can spread because culture is not acquired just from
parents.
There are good reasons to expect maladaptive beliefs to spread. As you know, we
acquire all our genes from our parents. But we acquire culturally transmitted beliefs and
values from our parents, friends, teachers, religious figures, the media, and other
sources. Biologist Richard Dawkins of the University of Oxford refers to culturally
transmitted beliefs and values as memes. The logic of natural selection applies to memes
in the same way that it does to genes. Memes compete for our memory and for our
attention, and not all memes can survive. Some memes are more likely to survive and
be transmitted than others. Memes are heritable, often passing from one individual to
another without major change. As a result, some memes spread, and others are lost.
Even though memes are subject to natural selection, it is not necessarily true that
only memes that enhance our genetic fitness will spread. The rules of cultural trans-
mission are different from the rules of genetic transmission, so the outcome of selec-
tion on memes can be different from the outcome of selection on genes. The basic rules
of genetic transmission are simple. With some exceptions, every gene that an individ-
ual carries in her body is equally likely to be incorporated into her gametes, and the
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 441
only way that those genes can be transmitted is through her offspring. Thus,
only genes that increase reproductive success will spread. Cultural trans-
mission is much more complicated. Memes are acquired and transmitted
throughout an individual’s life, and they can be acquired from grandpar-
ents, siblings, friends, coworkers, teachers, and even completely impersonal
sources such as books, television, and now the Internet.
Genes are transmitted directly from parents to offspring, but memes
may spread along many different pathways. This means that memes can
spread even if they do not enhance reproductive fitness. If ideas about dan-
gerous hobbies like rock climbing or heroin use spread from friend to friend,
these ideas can persist even though they reduce survival and individual repro-
ductive success (Figure 15.21). Beliefs about heaven and hell can spread from
priest to parishioner, even if the priest is celibate and thus has no offspring
of his own to influence. Moreover, cultural variants may accumulate and be
transmitted within groups of people who form clans, fraternities, business
firms, religious sects, or political parties. This process can generate groups
that are defined by cultural values and traditions, not by genetic relatedness.
The fact that culture can lead to outcomes not predicted by conven- Cultural evolution may
FIGURE 15.21
tional evolutionary theory does not mean that human behavior has some- permit the spread of
how transcended biology. The idea that culture is separate from biology is ideas and behaviors that do not contribute to repro-
a popular misconception that cannot withstand scrutiny. Culture is gener- ductive success. Dangerous sports, such as rock
ated from organic structures in the brain that were produced by the climbing, may be examples of such behaviors.
processes of evolution. However, cultural transmission leads to novel evo-
lutionary processes. Thus, to understand the whole of human behavior, evolutionary
theory must be modified to account for the complexities introduced by these poorly
understood processes.
The fact that culture can lead to outcomes that would not be predicted by con-
ventional evolutionary theory does not mean that ordinary evolutionary reasoning is
useless. The fact that there are processes that lead to the spread of risky behaviors like
rock climbing does not mean that these are the only processes that influence cultural
behavior. In this chapter, we saw that it is likely that many aspects of human psy-
chology have been shaped by natural selection so that people learn to behave adap-
tively. We love our children, have strong aversions to mating with close relatives, and
have an uncanny facility for learning language. There is every reason to suspect that
these predispositions play an important role in determining which memes spread and
which don’t. As long as this is the case, ordinary evolutionary reasoning will be use-
ful for understanding human behavior.
FURTHER READING
Barkow, J. H., L. Cosmides, and J. Tooby, eds. 1995. The Adapted Mind: Evolution-
ary Psychology and the Generation of Culture. New York: Oxford University
Press.
Fragaszy, D. M., and S. Perry, eds. 2003. The Biology of Traditions: Models and Evi-
dence. New York: Cambridge University Press.
15_Evolution [3p]:How Humans Evolved [5e] 10/6/08 4:37 PM Page 442
Richerson, P. J., and R. Boyd. 2005. Not by Genes Alone: How Culture Transformed
Human Evolution. Chicago: University of Chicago Press.
Whiten, A. V. Horner, and S. Marshall-Pescini. 2003. Cultural anthropology. Evolu-
tionary Anthropology 12:92–105.
KEY TERMS
Chapter 16
The Psychology of Human Mate Preferences
Some Social Consequences
of Mate Preferences
Kipsigis Bridewealth
Nyinba Polyandry
Raising Children
Parenting Effort and Mating Effort
Grandparental Care
Discriminative Parental Solicitude
Prenatal Investment
Infanticide
Adoption
Adoption in Oceania
Adoption in Industrialized Societies
Is Human Evolution Over?
Y
ou should be convinced by now that every aspect of the human phenotype
is the product of evolutionary processes. Over the last 10 million years, nat-
ural selection, mutation, and genetic drift have made us into large, nearly
hairless, bipedal primates with grasping hands and large brains. The same evolu-
tionary processes have molded the psychological mechanisms that influence human
behavior, causing us to behave differently from other primates in some contexts. There
is simply no other explanation.
443
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 444
However, the fact that modern humans are the product of evolutionary processes
does not necessarily mean that evolutionary theory will help us understand contem-
porary human behavior. Constraints on adaptation, which we discussed in Chapter 3,
may have predominated in the evolution of human behavior, limiting the usefulness
of adaptive reasoning. In Chapter 15, however, we saw that evolutionary theory has
provided useful insights into certain aspects of human psychology, social organization,
and culture. In this chapter, we will present several more examples that show how evo-
lutionary thinking can help us understand the day-to-day behavior of humans in mod-
ern societies.
In choosing these examples, we have focused on reproductive behavior because
mating and parenting strongly affect fitness. The unfortunate person who chooses a
mate that is lazy, infertile, or unfaithful is likely to have many fewer children than the
person whose spouse is healthy, hardworking, and faithful. Neglectful parents who fail
to provide adequate care for their offspring will raise fewer children than will atten-
tive parents who carefully nurture their offspring. Because reproductive decisions are
likely to have a marked effect on fitness, there is good reason to expect that our psy-
chology has been shaped by natural selection to improve the chances of making good
choices about who we mate with, who we marry, and how we raise our children. We
do not have enough space to provide a comprehensive analysis of these topics.
Instead, we will focus on certain examples in which evolutionary theory provides novel
and fundamental insights into human behavior and physiology.
These are the thoughts of 29-year-old Charles Darwin, recently returned from his five-
year voyage on the HMS Beagle. Soon after writing these words, Darwin married his
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 445
cousin Emma, the daughter of Josiah Wedgwood, the progressive and immensely
wealthy manufacturer of Wedgwood china (Figure 16.1). By all accounts, Charles and
Emma were a devoted couple. Emma bore Charles 10 children and nursed him
through countless bouts of illness. Charles toiled over his work and astutely managed
his investments, parlaying his modest inheritance and his wife’s more substantial one
into a considerable fortune.
Darwin’s frank reflections on advantages and disadvantages of marriage were very
much those of a conventional, upper-class Victorian gentleman. But people of every
culture, class, and gender have faced the problem of choosing mates. Sometimes peo-
ple choose their own mates, and other times parents arrange their children’s marriages.
But everywhere, people care about the kind of person they will marry.
For much of their evolutionary history, humans have lived in foraging societies.
Charles Darwin
The adaptive challenges that men and women face in these kinds of societies are likely FIGURE 16.1
courted and
to have shaped their mating strategies. For women, this might have meant choosing
married his cousin Emma Wedgwood.
men who would provide them with access to resources. Recall from Chapter 11 that
This portrait was painted when Emma
there is considerable interdependence between men and women in foraging societies. was 32, just after the birth of her first
Women mainly gather plant foods, and men are mainly responsible for hunting. child.
Women’s consumption exceeds their production for much of their reproductive life.
Children do not begin to provide substantial amounts of their own food until they
reach adolescence. Thus, for women, it might be important to choose a mate who will
be a good provider.
Men’s reproductive success depends largely on the fertility of their partners, so it
is plausible that selection favored men who focused on this attribute. Women’s fertil-
ity is highest when they are in their 20s and declines to zero when women reach
menopause, at about 50 years of age (Figure 16.2). Thus, selection should have favored
men who chose young and healthy mates. Because picture IDs were scarce in the Pleis-
tocene, selection may have shaped men’s psychology so that they are attracted to cues
that reliably predict youth and health, such as smooth skin, good muscle tone, sym-
metrical features, and shiny hair.
For both men and women, it is important to find mates that they can get along
with. Human children are dependent on their parents for a remarkably long time. Dur-
0.25
Men
Women
0.20
Age-specific fertility rate
0.15
People in 33
FIGURE 16.3
countries (red)
were surveyed about the qualities of an
ideal mate.
ing this period, both parents provide food and shelter for their children. Because
parental investment lasts for many years, adaptive thinking predicts that both men and
women will value traits in their partners that help them sustain their relationships.
Both are likely to value personal qualities like compatibility, agreeableness, reliability,
and tolerance.
David Buss, a psychologist at the University of Texas at Austin, was among the
first to test the evolutionary logic underlying human mating preferences and tactics.
Buss enlisted colleagues in 33 countries to administer to more than 10,000 men and
women standardized questionnaires about the qualities of desirable mates. Most of the
data were collected in Western industrialized nations, and most samples represent uni-
versity students in urban populations within those countries (Figure 16.3). In the ques-
tionnaires, people were asked to rate a number of traits of potential mates—good
looks, good financial prospects, compatibility, and so on—according to their desir-
ability. Respondents were also asked about their preferred age at marriage and the pre-
ferred age difference between themselves and their spouse.
People generally care most about the personal qualities of their mates.
Men and women around the world rate mutual attraction or love above all other
traits (Table 16.1). The next most highly desired traits for both men and women are
personal attributes, such as dependability, emotional stability and maturity, and a pleas-
ing disposition. Good health is the fifth most highly rated trait for men and the seventh
for women. Good financial prospect is the thirteenth most highly rated trait by men and
the twelfth by women. Good looks are rated tenth by men and thirteenth by women.
It is interesting, and somewhat surprising, that neither sex seems to value chastity
highly. Perhaps this is because people were asked to evaluate the desirability of sexual
experience before marriage (that is, virginity), not fidelity during their marriage.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 447
TABLE 16.1 Men and women from more than 33 countries around the world (shown in
Figure 16.3) were asked to rate the desirability of a variety of traits in
prospective mates. The rankings of the values assigned to each trait, on average, are given here.
Subjects were asked to rate each trait from 0 (irrelevant or unimportant) to 3 (indispensable).
Thus, high ranks (low numbers) represent traits that were generally thought to be important.
(From Table 4, in D. M. Buss et al., 1990, “International Preferences in Selecting Mates: A Study
of 37 Cultures,” Journal of Cross-Cultural Psychology 21: 5–47.)
Mutual attraction/love 1 1
Dependable character 2 2
Emotional stability and maturity 3 3
Pleasing disposition 4 4
Good health 5 7
Education and intelligence 6 5
Sociability 7 6
Desire for home and children 8 8
Refinement, neatness 9 10
Good looks 10 13
Ambition and industriousness 11 9
Good cook and housekeeper 12 15
Good financial prospect 13 12
Similar education 14 11
Favorable social status or rating 15 14
Chastity* 16 18
Similar religious background 17 16
Similar political background 18 17
*Chastity was defined in this study as having no sexual experience before marriage.
Even though the ranking of the scores assigned to these traits is similar for men
and women, there are consistent differences between men and women in how desir-
able each gender thinks these traits are. Buss found that people’s sex had the greatest
effect on their ratings of the following traits: “good financial prospect,” “good
looks,” “good cook and housekeeper,” “ambition and industriousness.” As the evo-
lutionary model predicts, women value good financial prospects and ambition more
than men do, and men value good looks more than women do. Gender has a smaller
and somewhat less uniform effect on ratings of chastity. In 23 populations, men value
chastity significantly more than women do; in the remaining populations, men and
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 448
women value chastity equally. There are no populations in which women value
chastity significantly more than men do.
Men and women differ about the preferred ages of their partners.
Evolutionary reasoning suggests that men’s mate preferences will be strongly influ-
enced by the reproductive potential of prospective mates. Therefore, we would predict
men to choose mates with high fertility or high reproductive value. By the same token,
we would expect women to be less concerned about their partner’s age than about their
partner’s ability to provide resources for them and their offspring.
A considerable amount of evidence suggests that men consistently seek and
marry partners who are younger than they are and that women seek and marry part-
ners somewhat older than themselves. Douglas Kenrick of Arizona State University
and Richard Keefe of Scottsdale Community College surveyed marriage records in two
cities in the United States and in a small Philippine village, as well as personal adver-
tisements in the United States, northern Europe, and India. In all of these cases, they
found a similar pattern. As men get older, the age difference between them and their
wives increases. Thus, young men marry women slightly younger than themselves, and
older men marry partners considerably younger than themselves (Figure 16.4a). As
women get older, there is little change in the age difference between them and their hus-
bands (Figure 16.4b). Newspaper advertisements in which advertisers specify the range
of ages for prospective mates demonstrate a very similar pattern. Henry Harpending
of the University of Utah has found very similar patterns among the Herero, a pas-
toralist group in the northern Kalahari Desert of Botswana, in which marriages are
unstable, divorce is common, and women have a considerable amount of financial
independence.
Although men advertise for and marry progressively younger women, the actual
age of their partners does not seem to fit the prediction that males will choose fertile
mates. Older men seek and marry women who are considerably younger than they
themselves are, but not young enough to produce children. Men’s choices about who
to date and who to marry may be driven by multiple factors, not just women’s fertil-
ity. Older men may desire younger women, but they may also want to find someone
who shares their taste in music, has similar goals in life, and so on. Furthermore, men’s
preferences and their marriages may reflect their own attractiveness in the mating mar-
ket. Older men may want young women but know they will have to settle for part-
ners closer to their own age. Together, the data taken from personal advertisements
and from marriage records reflect individual desires tempered with reality.
In addition to having different criteria for the ideal mate, men and women may
also have different mating tactics. Because women devote nine months to each preg-
nancy and nurse their children for even longer, selection is likely to have favored a psy-
chology that makes them cautious about involvement in sexual relationships that
would expose them to the risks of pregnancy. (Of course, nowadays birth control
reduces the risk of pregnancy for women, but effective methods of contraception are
a recent innovation. Human mating tactics evolved in a world without such technol-
ogy.) Women are likely to prefer stable, committed relationships with men who are
willing and able to help care for them and their offspring. Because the costs of con-
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 449
15
10
5
Age difference
0
Maximum
–5
Minimum
–10
–15
–20
<20 20–29 30–39 40–49 50–59 60–69
Male’s age
(a) Men’s preferences
15
10 Maximum
5
Age difference
0
Minimum
–5
ception are borne mainly by women, men can afford to be more flexible in their mat-
ing tactics, and to have a psychology that makes them more open to mating oppor-
tunities that do not involve long-term commitments. However, we would expect men
to form committed long-term relationships because children that receive care from
both parents are more likely to thrive.
David Schmitt of Bradley University has coordinated a comprehensive cross-
cultural study of human sexuality, sampling people, mainly university students, in 62
countries around the world. In this survey, people were asked about the traits that they
valued in potential mates and were also asked about various aspects of their mating tac-
tics. For example, they were asked the number of sexual partners they would like to have
over various time intervals ranging from 6 months to 30 years. For all time intervals,
men reported preferring a larger number of sexual partners than women did (Figure
16.5). This difference seems to be common cross-culturally (Figure 16.6), although the
magnitude of the sex difference and the number of partners desired vary considerably.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 450
20
18 Men
Women
16
12
10
When you meet someone you’re attracted to, you probably feel excitement and
some degree of uncertainty. Some of this uncertainty arises because you are not sure
whether the other person is as attracted to you as you are to him or her. And some of
this uncertainty arises because you don’t know what the other person’s intentions are.
Martie Haselton of the University of California, Los Angeles and David Buss have
pointed out that this uncertainty generates different kinds of problems for men and
women. To understand the logic of their argument, let’s think about what kinds of mis-
takes you could make in deciding whether someone was attracted to you. A false pos-
itive arises if you think the other person is attracted to you, when in fact that’s not the
case. A false negative occurs if you think the other person doesn’t like you, when he
7 Men
Women
6
Number of partners desired
1
In every world region,
FIGURE 16.6
men prefer signifi- 0
North South Western Eastern Southern Middle Africa Oceania South, East Asia
cantly more sexual partners than women do America America Europe Europe Europe East Southeast
over the next month. Asia
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 451
or she really does. Both kinds of errors are costly: a false positive could lead you to
make an overture that would be rejected (“Sorry, I need to wash my hair”); a false neg-
ative could prevent you from making any overture at all (and you will have nothing
better to do than wash your hair).
Now think about the kinds of errors that can arise when there is uncertainty about
the other person’s intentions about the relationship. Haselton and Buss hypothesize
that natural selection will predispose men and women to
bias their judgments about new partners’ sexual inten- Women’s estimate of
3.0 men’s commitment
tions and commitment in different ways. Women, who
Men’s estimate of
could become pregnant, are expected to be cautious about 2.5 own commitment
Likelihood of commitment
their partner’s intentions, and as a result they will make
more false negative errors than false positive errors. Put 2.0
another way, evolutionary reasoning predicts that women
are more likely to underestimate men’s commitment than 1.5
to overestimate it. Men, who are interested in pursuing
both short-term and long-term relationships, are expected 1.0
to minimize the chance of missing sexual opportunities,
0.5
and as a result they will make more false-positive errors
than false-negative errors. That is, they are likely to over-
0
estimate women’s sexual interest more often than they Passionately Said Held hands Invited to
kissed “I love you” apartment for
underestimate it.
drink
Haselton and Buss have conducted a number of dif- (a)
ferent studies on college students in the United States to
test this hypothesis. They have asked men and women to
Women’s estimate of own
evaluate sexual intent and commitment in members of sexual intent
their own sex and the opposite sex; to imagine how they 2.5 Men’s estimate of women’s
would interpret various kinds of signals directed to them- sexual intent
2.0
selves (for example, held hands, declared love); and to
recall instances when their own intentions were misun- 1.5
Sexual intent
16.7). – 0.5
–1.0
Culture predicts people’s mate preferences better –1.5
than gender does. Passionately
kissed
Said
“I love you”
Held hands Invited to
apartment for
drink
Both of the large cross-cultural data sets reveal con- (b)
siderable variation from country to country. Buss and his
colleagues found that the country of residence has a Men and women were asked to estimate how
FIGURE 16.7
greater effect than gender does on variation in all of the they would interpret particular courtship signals
by members of the opposite sex and how they would rate the same sig-
18 traits in Table 16.1, except for “good financial
nals by members of their own sex, using a scale from +3 (very likely) to
prospect.” This means that knowing where a person lives
– 3 (very unlikely). (a) Men and women estimated the extent of commit-
tells you more about what he or she values in a mate than
ment implied by a number of different courtship signals. Women tended
knowing the person’s gender does. Of the 18 traits, to underestimate men’s interest in commitment relative to men’s own esti-
chastity (defined in Buss’s study as no sexual experience mates of their interest in commitment. (b) Men and women estimated
before marriage) shows the greatest variability among the likelihood of sexual intent implied by the same signals. Men tended
populations. In Sweden, men rate chastity at 0.25, and to overestimate women’s sexual intent relative to women’s perceptions of
women rate it at 0.28 on a scale of 0 (irrelevant) to 3 their own intent.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 452
Women
Men
Sweden Sweden
Australia Australia
USA USA
Belgium Belgium
Bulgaria Bulgaria
China China
Nigeria Nigeria
0 0.5 1.0 1.5 2.0 2.5 3.0 0 0.5 1.0 1.5 2.0 2.5
Irrelevant Indispensable Irrelevant Indispensable
Average rating Average rating
(indispensable). In contrast, Chinese men rate chastity 2.54, and Chinese women rate
it 2.61 (Figure 16.8). This means that there is more similarity between men and women
from the same population than there is among members of each sex from different
populations.
This result illustrates an important point: evolutionary explanations that invoke
an evolved psychology and cultural explanations that are based on the social and cul-
tural milieu are not mutually exclusive. The cross-cultural data suggest there are some
uniformities in people’s mate preferences that are the result of evolved psychological
mechanisms. People everywhere want to marry kind, caring, trustworthy people. Men
want to marry young women and women want to marry older men. But this is not the
whole story. The cross-cultural data also suggest that human mate preferences are
strongly influenced by the cultural and economic environment in which we live. Ulti-
mately, culture also arises out of our evolved psychology, and the cultural variation in
mate preferences that the cross-cultural data reveal must, therefore, also be explica-
ble in evolutionary terms. However, the way our evolved psychology shapes the cul-
tures in which we live is complicated and poorly understood, and many interesting
questions remain unresolved. Evolutionary theory does not yet explain, for example,
why chastity is considered essential in China but undesirable in Sweden.
cultural values, which celebrate women’s youth and beauty and men’s wealth and
power. They argue that researchers are not studying evolved preferences, but rather
are learning about cultural values and beliefs. In response, advocates of evolutionary
analyses argue that the cross-cultural uniformity of mate preferences and mating tac-
tics reflects evolved psychological predispositions that are modified, but not created,
by culture.
Other critics have accepted the general logic of evolutionary reasoning but have
questioned the methods used to assess mate preferences and mating tactics. Most of
the early work was based on pencil-and-paper tests in which people, often under-
graduates, were asked about their preferences and their personal experiences. These
kinds of data may be biased in a number of ways. For example, cultural norms may
lead men to exaggerate their sexual experience and women to understate their desire
for sexual variety, even on anonymous surveys.
Moreover, for reasons that are not fully understood, results derived from “forced
choice” tests, in which subjects are asked which of two alternatives they would choose,
sometimes deviate from results in which subjects are asked to rate the same alterna-
tives. Although this may seem like an arcane methodological issue, it may be impor-
tant if different kinds of data generate different conclusions about the psychology
underlying mating tactics.
This problem arises in studies of sex differences in jealousy. Buss and his colleagues
have hypothesized that there will be sex differences in responses to emotional and sex-
ual infidelity. Men, who are concerned about being cuckolded, will be more upset
about their partners’ sexual infidelity than about their emotional infidelity. On the
other hand, women, who are concerned about access to resources for themselves and
their offspring, will be more upset about their partners’ emotional infidelity than about
their sexual infidelity. When subjects are asked which form of infidelity would upset
them more, men typically report that they would be more upset by sexual infidelity
than emotional infidelity, and women report the opposite pattern. However, when
Christine Harris of the University of California, San Diego asked men and women how
much they focused on each type of infidelity, these sex differences disappeared.
Harris suggests that the forced-choice paradigm provides a misleading picture of
the sex differences underlying jealousy. It’s possible, she argues, that men find sexual
infidelity more upsetting than emotional infidelity in the forced-choice paradigm
because they assume that their partners would not sleep with someone else unless they
were emotionally involved with them; thus, sexual infidelity is more upsetting because
it implies both emotional and sexual betrayal. Similarly, women may be more upset
by emotional infidelity than by sexual infidelity because they assume that if their part-
ners are emotionally involved with someone else, they are also sleeping with them, but
not vice versa. Although this explanation is cogent, we don’t know if it is correct.
It is not yet clear whether sex differences in sexual jealousy are robust or which
type of methodology generates the most meaningful information. However, researchers
are developing a battery of new methods that will allow them to investigate sexual jeal-
ousy and other aspects of mating preferences more fully.
people’s behavior in social situations and, consequently, shape the societies in which
they live.
Kipsigis Bridewealth
Among the Kipsigis, a group of Kalenjin-speaking people who live in the Rift Valley
province of Kenya, women usually marry in their late teens (Figure 16.9), men usu-
ally marry for the first time when they are in their early 20s,
and it is common for men to have several wives—a practice
called polygyny. As in many other societies, the groom’s
father makes a bridewealth payment to the father of the
bride at the time of marriage. The payment, tendered in live-
stock and cash, compensates the bride’s family for the loss
of her labor and gives the groom rights to her labor and the
children she bears during her marriage. The amount of the
payment is settled through protracted negotiations between
the father of the groom and the father of the bride. The
average bridewealth consists of six cows, six goats or sheep,
and 800 Kenyan shillings. This is about one-third of the
average man’s cattle holdings, one-half of his goat and
sheep herd, and two months’ wages for men who hold
These Kipsigis women are eligible for mar- salaried positions. Because men marry polygynously, there
FIGURE 16.9
riage. Their fathers will negotiate with the is competition over eligible women. Often the bride’s father
fathers of their prospective husbands over bridewealth. (Photograph entertains several competing marriage offers before he
courtesy of Monique Borgerhoff Mulder.) chooses a groom for his daughter. The prospective bride and
groom have little voice in the decisions their fathers make.
Anthropologist Monique Borgerhoff Mulder at the University of California, Davis
reasoned that Kipsigis bridewealth payments provide a concrete index of the qualities
that each party values in prospective spouses. The groom’s father is likely to prefer a
bride who will bear his son many healthy children. His bridewealth offer is expected
to reflect the potential reproductive value of the prospective bride. The groom’s father
is also expected to prefer that his son marry a woman who will devote her labor to his
household. Kipsigis women who remain near their own family’s households are likely
to be called on to help their mothers with the harvest and to assist their mothers in child-
birth. Thus, the groom’s father may prefer a woman whose natal family is distant from
his son’s household. The bride’s father is likely to have a different perspective on the
negotiations. Because wealthy men can provide their wives with larger plots of land to
farm and more resources, we would expect the bride’s father to prefer that his daugh-
ter marry a relatively wealthy man. At the same time, because the bride’s family will be
deprived of her labor and assistance if she moves far away from their land, the bride’s
father is likely to prefer a groom who lives nearby. The fathers of the bride and groom
are expected to weigh the costs and benefits of prospective unions in their negotiations
over bridewealth payments. For example, although the bride’s father may prefer a high
bridewealth payment, he may settle for a lower payment if the groom is particularly
desirable. In order to determine whether these preferences affected bridewealth pay-
ments, Borgerhoff Mulder recorded the number of cows, sheep, and goats, as well as
the amount of money, that each groom’s family paid to the bride’s family.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:19 PM Page 455
Plump women whose menarche occurred at an early age fetched the highest
bridewealth payments.
Borgerhoff Mulder found that bridewealth increased as the bride’s age at the time
of menarche (her first menstruation) decreased (Figure 16.10). That is, the highest
bridewealths were paid for the women who were youngest when they first menstru-
ated. Among the Kipsigis, age at menarche is a reliable index of women’s reproduc-
tive potential. Kipsigis women who reach menarche early have longer reproductive life
spans, higher annual fertility, and higher survivorship among their offspring than do
women who mature at later ages.
Borgerhoff Mulder wondered how a man assessed his prospective bride’s repro-
ductive potential, since men often do not know their bride’s exact age, nor her age at
reaching menarche. One way to be sure of a woman’s ability to produce children would
be to select one who had already demonstrated her fertility by becoming pregnant or
producing a child. However, bridewealths for such women were typically lower than
bridewealths for women who had never conceived. Instead, bridewealth payments
were associated with the physical attributes of women. The fathers of brides who were
considered by the Kipsigis to be plump received significantly higher bridewealth pay-
ments than did the fathers of brides considered to be skinny. The plumpness of prospec-
tive brides may be a reliable correlate of the age at menarche, since menarcheal age is
determined partly by body weight. Plumpness may also be valued because a woman’s
ability to conceive is related to her nutritional status.
Bridewealth payments are also related to the distance between the bride’s home
and the groom’s home; the farther she moves, the less likely she is to provide help to
her mother, and the higher the bridewealth payment will be. However, there is no rela-
tionship between the wealth of the father of the groom and the bridewealth payment.
The bride’s father does not lower the bridewealth payment to secure a wealthy hus-
band for his daughter. Although this finding was unexpected, Borgerhoff Mulder sug-
gested that it may be related to the fact that differences in wealth among the Kipsigis
are unstable over time. A wealthy man who has large livestock herds may become
12
9
Bridewealth units
Kipsigis girls
FIGURE 16.10
6 who mature
early fetch larger bridewealth payments
than older girls do. Among the Kipsigis,
girls undergo circumcision (removal of
3 the clitoris) within a year of menarche.
The largest bridewealths were paid for
the girls who underwent menarche and
circumcision at the youngest ages.
Bridewealth is transformed into standard-
12 13 14 15 16 17 18 ized units to account for the fact that the
Age at circumcision value of livestock varies over time.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 456
relatively poor if his herds are raided, decimated by disease, or diverted to pay for
another wife. Although land is not subject to these vicissitudes, the Kipsigis tradi-
tionally have not held legal title to their lands.
Nyinba Polyandry
Because mammalian females are generally the limiting resource for males, males usu-
ally compete for access to females. As a result, among mammalian species polygyny is
much more common than polyandry. As you may recall from Chapter 5, polyandry
is a mating system in which one female is paired with two or more males. Polyandry
is rare because, all other things being equal, males that share access to a single female
produce fewer offspring than do males that maintain exclusive access to one or more
females. The reproductive costs of polyandry may be somewhat reduced if brothers
share access to a female—an arrangement called fraternal polyandry. Thus, in accor-
dance with Hamilton’s rule (see Chapter 7), it is plausible to assume that natural selec-
tion has shaped human psychology so that a man is more willing to care for his
brother’s offspring than for children who are unrelated to him.
Polyandrous marriage systems are as rare among human societies as they are in
other mammalian species. Anthropologist George Peter Murdock compiled informa-
tion about marriage practices in a sample of 862 societies around the world. In 83%
of these societies, polygyny is the preferred form of marriage. In 16% of the societies,
marriages are exclusively monogamous. Polyandry is practiced in only 0.5% of Mur-
dock’s sample.
Although polyandry is generally rare among human societies, there are several
societies in the highlands of the Himalayas in which
fraternal polyandry is the preferred form of marriage.
In these societies, several brothers marry one woman
and establish a communal household. Because
polyandry seems to limit a man’s reproductive oppor-
tunities, we might ask why men tolerate polyandrous
marriages. Nancy Levine of the University of Califor-
nia, Los Angeles studied one of these polyandrous
societies: the Nyinba of northwestern Nepal (Figure
16.11). Her data shed some interesting light on this
question.
The Nyinba live in four prosperous villages, nes-
tled between 2,850 and 3,300 m (about 9,500 and
11,000 ft) on gently sloping hillsides. Nyinba families
support themselves through a combination of agricul-
Polyandry is the preferred form of marriage among ture, herding, and long-distance trade. The Nyinba
FIGURE 16.11 believe that marriages of three brothers are most desir-
the Nyinba. In this family, three brothers are mar-
ried to one woman. The Nyinba consider this to be the ideal number of able, allowing one husband to farm, another to herd
cohusbands. (Photograph courtesy of Nancy Levine.) livestock, and the third to engage in trade. In practice,
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 457
however, all brothers marry jointly, no matter how many there are. The Nyinba are
quite concerned with the paternity of their children. Women identify the fathers of
each of their children, a task made easier by the fact that one or more of their hus-
bands is often away from home for lengthy periods tending to family business.
Although it is impossible to be certain how accurate these paternity assessments are,
it is clear that the Nyinba place great stock in them. Men develop particularly close
relationships with the children they have fathered, and fathers bequeath their share
in the family’s landholdings to their own sons.
Polyandry is the ideal form of marriage among the Nyinba, but reality does not
always conform to this ideal. Although all men marry jointly, some marriages break
down when one or more of the brothers brings another wife into the household or
leaves to set up an independent household. When households break up, each man
receives a share of the estate and takes with him all the children he has fathered dur-
ing the marriage. Levine reasoned that a comparison of marriages that remain intact
and marriages that break up would provide clues about the kinds of problems that
arise in polyandrous marriages. Together with one of us (J.B.S.), Levine has used her
data to determine whether male decisions conform to predictions derived from evo-
lutionary theory.
All other things being equal, the more men who are married to a single woman,
the lower each man’s reproductive success is likely to be. Thus, we might expect mar-
riages with many cohusbands to be less stable than marriages with fewer cohusbands.
This turns out to be the case; few of the largest communal Nyinba marriages remained
intact. Only 10% of the marriages that began with two cohusbands partitioned (split
apart), but 58% of the marriages that began with four cohusbands did (Figure 16.12).
Some Nyinba marriages are arranged by parents; others are set up by the oldest
brother. In either case, the wife is usually a few years younger than the oldest cohus-
40
Partition
Conjoint
Polyandrous
30
Number of households
20
band. In some marriages, the youngest cohusbands are considerably younger than their
wives. Remember that Buss’s cross-cultural data suggest that men prefer to marry
younger women. Thus, we might expect that men would be dissatisfied with marriages
to women much older than themselves. In fact, the Nyinba men who were most jun-
ior to their wives were the ones most likely to initiate partitions. When these men
remarried, they invariably chose women who were younger than themselves and who
were younger than their first wives.
If the basic problem with polyandry is that it limits male reproductive opportu-
nities, then we should expect males’ satisfaction with their marriages to be linked to
their reproductive performance. In most marriages, the oldest brother fathers the wife’s
first child. The second-oldest brother often fathers the second child, and so on. How-
ever, disparities in reproductive success among cohusbands often persist. A man’s place
in the birth order is directly related to his reproductive success, so the oldest brothers
have more children than the youngest brothers have. Men’s decisions to terminate their
marriages are associated with their reproductive performance: men who remained in
stable polyandrous marriages fathered an average of 0.1 children per year; men who
terminated their marriages had produced, on average, only 0.04 children per year dur-
ing the relationship.
Finally, kin selection theory would lead us to predict that close kinship among
cohusbands would help stabilize polyandrous marriages because men should be more
willing to invest in their brothers’ children. We can test this prediction because there
is a considerable amount of variation in the degrees of relatedness among cohusbands
within households. Some households are composed entirely of full siblings. In others,
the cohusbands may have multiple fathers, who are related themselves, and multiple
mothers, who may also be related. The result is a complicated web of relationships
among cohusbands. Contrary to predictions based on kin selection, there is no differ-
ence in the degree of relatedness among cohusbands in households that remained intact
and households that dissolved. Moreover, when men remarried, they showed no incli-
nation to align themselves with the cohusbands to whom they were more closely related.
RAISING CHILDREN
Children are a lot of work. In addition to the maternal burdens of pregnancy and lac-
tation, children need to be fed, carried, protected, clothed, and educated. In most
human societies, this job is done mainly by parents. All this seems so normal and so
obvious that it is easy to overlook the possibility that societies could be organized in
many other ways. For example, babies could be bought and sold or traded back and
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 459
forth like pets or domesticated animals. Such a system would have lots of practical
advantages. It would be easy to adjust the size of your family, to select the ratio or
order of boys and girls, and to regulate the spacing between children. Women who pre-
ferred to avoid pregnancy and delivery or who were unable to bear children could have
a family just as easily as anyone else. Some people might even opt to skip the dubious
pleasures of frequent diaper changes and midnight feedings, and acquire children who
were already two or three years old. But this is not what happens. Instead, in virtu-
ally every society, most people raise their own children. It seems likely that people do
this because their evolved psychology causes them to value their own children very dif-
ferently from other people’s children.
There is, however, considerable cross-cultural variation in human parenting
behavior. In some societies, most men are devoted fathers; in others, men take little
interest in children and rarely interact with their own offspring. In some societies, par-
ents supervise their children’s every waking moment; in others, children are reared
mainly by older siblings and cousins. Moreover, some aspects of parenting behavior
in some societies seem hard to reconcile with the idea that human psychology has been
shaped by natural selection. There are cultures in which parents sometimes kill or
physically abuse their own infant children, and cultures in which people regularly raise
children who are not their own. Because such behaviors seem inconsistent with the
idea that human behavior is controlled by evolved predispositions, their existence has
been cited as evidence that evolutionary reasoning has little relevance to understand-
ing modern human behavior.
In the remainder of this chapter, we will consider data suggesting that observed
patterns of parental investment, forms of neglect and abuse of children, and adoption
are consistent with the idea that human behavior is influenced by evolved predisposi-
tions. Moreover, evolutionary reasoning can give fresh and useful insights about when
and why such behaviors occur. However, we will also see that there are aspects of con-
temporary parenting behavior that evolutionary reasoning has not yet explained.
Some human paternal investment is parenting effort, but some is mating effort.
Human children require care from their parents long after they are weaned. The lengthy
period of parental investment means that some marriages will end before children are
grown. Although death, separation, and divorce divide families, remarriage can create
new families that include children from the partners’ previous marriages and children
from their present marriage. The theory of kin selection generally predicts that men and
women will direct care selectively to kin, and this might lead them to favor genetic off-
spring over stepchildren. At the same time, if men and women value their mates’ invest-
ment in their children, then care for children might also be a form of mating effort.
Kermyt Anderson, Hillard Kaplan, and Jane Lancaster have pointed out that, in
humans, parental care may function as both parenting effort and mating effort. Thus,
when husbands and wives invest in their genetic offspring, their efforts may enhance
their inclusive fitness and the quality and durability of their marital relationship.
Although parenting effort and mating are confounded when we think about husbands
and wives investing in their joint progeny, it is possible to tease mating effort and par-
enting effort apart, as we show in Table 16.2. As you can see, investment in stepchildren
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 460
TABLE 16.2 When men invest time, energy, or other resources in children, they may be
enhancing their parenting effort or their mating effort.
reflects mating effort but not parenting effort. On the other hand, investment in one’s
own offspring from a previous marriage reflects parenting effort but not mating effort.
Investment in a mate’s stepchildren from a previous relationship does not constitute
parenting effort or mating effort.
Anderson and his colleagues interviewed a large number of men in Albuquerque,
New Mexico, and asked them about their reproductive histories. They gathered infor-
mation about a number of different measures of parental care, including whether men
provided any financial support for children attending college and how much time they
spent with children in the previous year. The data in Figure 16.13 show that both mat-
ing effort and parenting effort seem to influence the extent of men’s investment in chil-
dren. Men are much more likely to provide money for their college-age children when
they are still married to the mothers of those children than they are to provide money
for their own children after the marriage has ended or for stepchildren. The magni-
tude of the effects of mating effort and parenting effort seem to be about the same; men
invest nearly as much in their stepchildren as they do in their own offspring from a pre-
vious relationship.
There is some evidence that similar patterns influence men in other societies.
Anderson and his colleagues surveyed Xhosa high-school students in Cape Town,
South Africa, and asked them about the amount and kinds of care they obtained from
0.80
0.75
Proportion of children receiving man’s support
0.70
0.65
0.60
0.55
0.50
0.45
The proportion of children attend-
FIGURE 16.13
ing college who receive support 0.40
from men varies considerably. Men are significantly more likely 0.35
to contribute to the costs of college for their own children
when they are still with the child’s mother than they are to 0.30
contribute to the costs of college for their stepchildren. How- 0.25
ever, men are equally likely to provide funds for stepchildren Man’s child Man’s stepchild Man’s child Wife’s stepchild
by current by previous by previous
and their own children by a previous mate. mate mate mate
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 461
600
their fathers and stepfathers. These students spent more time with their res-
ident genetic fathers than with their resident stepfathers, and they spent
more time with their resident stepfathers than with nonresident genetic
fathers (Figure 16.14). At the same time, the students generally received less
financial support from their stepfathers than from their genetic fathers; they
received the most support from resident genetic fathers.
Frank Marlowe, now at Florida State University, has investigated the
trade-off between mating effort and parenting effort among the Hadza, a
foraging group in Tanzania. Marriages among the Hadza are relatively
impermanent, so men may find themselves living with the mothers of their
children or with women who have children from previous relationships.
Hadza men hunt and bring food back to camp, where it is distributed.
Although they are out of camp for about the same amounts of time, men
who live with only biological offspring bring substantially more food back
to camp than do men who live with at least one stepchild.
Hadza men also spend about 2% to 5% of every day caring directly for
children. This care takes various forms: they may hold, clean, feed, com-
fort, or feed children; play with them; talk with them; and stay close to
them (Figure 16.15). Marlowe predicted that men would provide more care
for genetic offspring than for stepchildren, but he also predicted that men
would make trade-offs between the extent of their parenting effort and of
their mating effort. In addition, Marlowe postulated that when men’s mat-
ing opportunities were limited, they would invest more time in caring for
A Hadza man holds his
their genetic offspring; and that when mating opportunities were abundant, FIGURE 16.15
child on his lap. Men pro-
men would increase mating effort and decrease parenting effort. As he vide more care for genetic offspring than for
expected, Marlowe found that Hadza men provide more care for their stepchildren, and they seem to balance mating effort
genetic offspring than for stepchildren, but they do spend some time car- against parenting effort. (Photo courtesy of Frank
ing for stepchildren. Caring for stepchildren might be a form of mating Marlowe.)
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 462
effort. Moreover, when men were living in larger camps with more eligible women,
they provided less direct care for children.
Grandparental Care
In human societies, men and women often live long enough to have grandchildren.
Grandparents may play an active role in their grandchildren’s lives, particularly if all
their own children are grown. Grandparental care is likely to be a product of kin selec-
tion because grandparents are closely related to their grandchildren.
6.0 logic tells us that men are related to their sons and
daughters by 0.5p, to their daughters’ children by
5.5
0.25p, and to their sons’ children by 0.25p2. When-
ever p ⬍ 1, women will be more closely related to
5.0
their children than their husbands are, and both
4.5 parents will be more closely related to their daugh-
ters’ children than to their sons’ children.
4.0 This asymmetry in relatedness to grandchildren
led Harald Euler and Barbara Weitzel of the Univer-
3.5
Maternal Maternal Paternal Paternal
sity of Kassel to predict that there would be matri-
grandmother grandfather grandmother grandfather lateral biases in grandparental investment. Euler and
Weitzel surveyed nearly 2,000 German adults and
In Germany, maternal grandparents provide more
FIGURE 16.16 asked them how much care they received from their
care on average than paternal grandparents, perhaps
grandparents (Figure 16.16). In this population,
because parents are more confident of their daughters’ maternity than their
sons’ paternity. Adults were asked to rate the amount of care that they both maternal grandparents provide more care than
received from each of their grandparents on a seven-point scale. High scores their paternal counterparts do. And on both sides of
correspond to higher amounts of care. Here the mean and the standard devia- the family, grandmothers provide more care for chil-
tions are plotted. dren than grandfathers do. However, maternal
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 463
grandfathers provide significantly more care than paternal grandmothers do, even
though both are related to their grandchildren by 0.25p. Of course, the matrilateral
bias in care for grandchildren might arise if there were a tendency for children to live
closer to their maternal grandparents than to their paternal grandparents. Euler and
Weitzel’s data indicate that this is not the case, however. Moreover, the pattern held
when the authors controlled for grandparents’ age and availability.
Parents’ ability to invest in their offspring is limited, so investment in one offspring lim-
its their ability to invest in other offspring. If a child’s prospects for survival are poor
or the costs of raising a particular child are very high, then investment will not enhance
parental fitness. Thus, natural selection is expected to favor mechanisms that cause par-
ents to adjust investment in offspring in relation to the offspring’s impact on parental
fitness. Before birth, maternal physiology is expected to terminate pregnancies when
fetal prospects are poor or the costs of carrying the fetus to term threaten maternal
health. After birth, we expect parental psychology to be sensitive to the child’s condi-
tion, the parents’ economic circumstances, and the parents’ alternative opportunities
to pass on their genetic material. This phenomenon has been labeled discriminative
parental solicitude by psychologists Martin Daly and Margo Wilson of McMaster Uni-
versity in Ontario, Canada.
Prenatal Investment
The costs of pregnancy must be weighed against the potential benefits. For women,
every pregnancy is a costly investment. Pregnancy lasts nine months, and during their
pregnancies, women need more food and work somewhat less efficiently. Childbirth
is also a major source of mortality among women. The magnitude of these costs and
their impact on the mother’s future reproductive success will depend partly on the
mother’s condition at conception. Women who are in poor health or badly nourished
will be less able to meet the demands of pregnancy and will take longer to recover from
their pregnancies than will healthy, well-nourished women.
The fitness benefit that the mother will gain depends largely on the quality of the
fetus that she carries. If the fetus is sickly or suffers from a genetic defect, it will be
unlikely to survive. If the mother is in poor condition, she may be unable to provide
the resources that her fetus or infant needs to survive. In such cases, the costs of the
pregnancy will outweigh the benefits. Thus, we would expect natural selection to favor
physiological mechanisms that terminate a pregnancy when there is something wrong
with the fetus or the mother is in very poor condition.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 464
It has been estimated that only 22% of all conceptions are carried to term. The
rest are miscarried, usually before the twelfth week of pregnancy (Figure 16.17). In the
majority of cases, miscarriages are associated with overt chromosomal abnormalities
in the fetus. There are good reasons to expect that natural selection acting on moth-
ers would favor miscarriages under these circumstances.
Most people think of miscarriages as pathologies—maladaptive and unpre-
dictable aberrations that cause the fetus to die and the pregnancy to be terminated. For
those who suffer miscarriages, the loss of a fetus is often deeply distressing. In such
cases, parents may never know why their fetus died or whether their next one is likely
to survive. As upsetting as miscarriages are for parents, there is good reason to believe
that some of the mechanisms that produce miscarriages may be adaptive responses that
have been favored by natural selection over the course of human history.
There is conflict between mother and fetus over whether a given pregnancy
should be sustained.
To this point we have considered miscarriages from the point of view of selection
acting on the mother. The picture looks different from the perspective of the fetus. As
we explained in Chapter 6, the fetus has a greater interest in its own survival (and a
weaker interest in any future sibling’s survival) than does its mother. Thus, there will
be a certain scope for conflict between the mother and her fetus over whether a given
pregnancy should continue. Natural selection will favor genes expressed in the fetus
that prevent those spontaneous abortions that are induced by genes expressed in the
mother. The fetus plays an active role in promoting its own survival.
In humans, pregnancy is maintained by the production of luteinizing hormone
(LH) by the anterior pituitary gland. Luteinizing hormone, in turn, stimulates the pro-
duction of another hormone, progesterone, by the corpus luteum, a mass of endocrine
cells formed in the mother’s ovary after ovulation. Progesterone inhibits uterine con-
tractions that would terminate the pregnancy.
From very early in pregnancy, the fetus influences the levels of progesterone in its
mother’s bloodstream. Initially, the fetus releases human chorionic gonadotropin
(hCG), which stimulates the production of progesterone by the mother’s corpus
luteum, directly into the mother’s bloodstream. By the eighth week of pregnancy, how-
ever, the fetus can produce enough progesterone to sustain the pregnancy, even if the
corpus luteum is removed from the mother. Thus, the eight-week-old fetus is able to
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 465
manipulate the maternal endocrine system to enhance its own viability (Figure 16.18).
It seems plausible that the ability of the fetus to manipulate the level of progesterone
in its mother’s bloodstream has been favored by natural selection because this ability
allows the fetus to protect itself from premature termination of the pregnancy.
Infanticide
TABLE 16.3 In a representative cross-cultural sample of societies, the rationales for infan-
ticide are consistent with predictions derived from evolutionary theory. Par-
ents commit infanticide mainly when the child is unlikely to thrive, when their own circumstances
make it difficult to raise the child, or when the child was not fathered by the mother’s spouse.
(Data from Table 3.1, in M. Daly and M. Wilson, 1988, Homicide [New York: de Gruyter].)
Adoption
In many societies, children are adopted and raised by adults other than their
biological parents.
Adoption is the flip side of infanticide. In some regions of the world, particularly Ocea-
nia and the Arctic, a substantial fraction of all children are raised in adoptive house-
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 467
ADOPTION IN OCEANIA
No data
Unrelated 0.2500
Unknown r
<0.0313
0.0313
• Natural parents maintain contact with their children after adoption and terminate
adoptions if children are neglected, mistreated, or unhappy.
• Natural parents prefer to have their children adopted by well-to-do individuals
who can provide adequately for the children’s daily needs or bestow property on
them.
• Sometimes asymmetries in investment exist between natural and adoptive chil-
dren; adopted children often inherit less property from their adoptive parents than
biological offspring do.
As you read the preceding description of adoption in the traditional societies of Ocea-
nia, you may have been struck by how different these transactions are from adoptions
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 469
TABLE 16.4 The patterns of adoption in the United States and Oceania show both simi-
larities and differences.
*In the United States from 1952 to 1971, approximately one-fourth of adoptions involved kin,
another fourth involved stepparents, and one-half involved unrelated individuals.
in the United States. In the United States and in other industrialized nations, adoption
is often a formal, legal process involving strangers. But as you compare the features
of adoption transactions in Oceania and the United States in Table 16.4, you will see
that there are similarities as well as differences between them.
In Oceania and in the United States, children are given up for adoption mainly
when their parents are unable to raise them, and they are frequently adopted by peo-
ple who have no dependent children of their own or by people who can afford to raise
additional children. In both Oceania and the United States, biological parents often
give up their children with considerable reluctance and hope to place their children in
homes where the children’s prospects will be improved.
The main difference between adoption in Oceania and in the United States is that
in Oceania, adoption is a relatively informal, open transaction between close kin, while
in the United States, adoption is often a formal, legal transaction between strangers.
The anonymity of the U.S. model has a number of ramifications. When the identity of
the biological parents is not disclosed to the adoptive parents or the adopted child,
contact between biological parents and their children is broken. In Oceania, adopted
children’s interests are protected by their biological parents, who maintain contact with
them and reserve the right to terminate the adoption if the children are mistreated or
unhappy. In the United States, government agencies are responsible for protecting the
interests of adopted children. In recent years, some of these differences have been
eroded as open adoptions have become more common, and the courts have given
adoptees the right to find out the identities of their birth parents.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 470
The different patterns of adoption in Oceania and the United States may result
from the same basic evolved psychological motivations.
FURTHER READING
Barrett, L., R. Dunbar, and J. Lycett. 2002. Human Evolutionary Psychology. Prince-
ton: Princeton University Press.
16_Human Mate Choice [4p]:How Humans Evolved [5e] 10/10/08 12:20 PM Page 471
Betzig, L., M. Borgerhoff Mulder, and P. Turke, eds. 1988. Human Reproductive
Behavior: A Darwinian Perspective. New York: Cambridge University Press.
Buss, D. M. 2003. The Evolution of Desire: Strategies of Human Mating. Rev. ed. New
York: Basic Books.
Daly, M., and M. Wilson. 1988. Homicide. New York: de Gruyter.
Silk, J. B. 1990. Human adoption in evolutionary perspective. Human Nature 1:25–
52.
KEY TERMS
Epilogue
THERE IS GRANDEUR IN
THIS VIEW OF LIFE . . .
H
ere we end our account of how humans evolved. As we promised in the
Prologue, the story has not been a simple one. We began, in Part One, by
explaining how evolution works: how evolutionary processes create the
exquisite complexity of organic design, and how these processes give rise to the stun-
ning diversity of life. Next we used these ideas in Part Two to understand the ecol-
ogy and behavior of nonhuman primates: why they live in groups, why the behavior
of males and females differs, why animals compete and cooperate, and why primates
are so smart compared with other kinds of animals. Then, in Part Three, we combined
our understanding of how evolution works and our knowledge of the behavior of
473
17_Epilogue [3p]:How Humans Evolved [5e] 10/6/08 4:45 PM Page 474
474 Epilogue
other primates with information gleaned from the fossil record to reconstruct the his-
tory of the human lineage. We traced each step in the transformation from a shrew-
like insectivore living at the time of dinosaurs; to a monkeylike creature inhabiting the
Oligocene swamps of northern Africa; to an apelike creature living in the canopy of
the Miocene forests; to the small-brained, bipedal hominins who ranged over Pliocene
woodlands and savannas; to the large-brained and technically more skilled early mem-
bers of the genus Homo, who migrated to most of the Old World; and, finally, to crea-
tures much like ourselves who created spectacular art, constructed simple structures,
and hunted large and dangerous game just 100 kya. Finally, in Part Four, we turned
to look at ourselves—to assess the magnitude and significance of genetic variation in
the human species, to ask how culture influences human evolution, and to consider
how we choose our mates and how we raise our children.
Evolutionary analyses of human behavior are not always well received. In Dar-
win’s day, many were deeply troubled by the implications of his theory. One Victorian
matron, informed that Darwin believed humans to be descended from apes, is
reported to have said, “Let us hope that it is not true, and if it is true, that it does not
become widely known.” Darwin’s theory profoundly changed the way we see our-
selves. Before Darwin, most people believed that humans were fundamentally differ-
ent from other animals. Human uniqueness and human superiority were unquestioned.
But we now know that all aspects of the human phenotype are products of organic
evolution—exactly the same processes that create the diversity of life around us.
Nonetheless, many people still feel that we diminish ourselves by explaining human
behavior in the same terms that we use to explain the behavior of chimpanzees or
soapberry bugs or finches.
Charles Darwin died In contrast, we think the story of human evolution is breathtaking in its grandeur.
FIGURE 1 With a few simple processes, we can explain how we arose, why we are the way we
in 1882 and was
buried in Westminster Abbey beneath are, and how we relate to the rest of the universe. It is an amazing story. But perhaps
the monument to Isaac Newton. Darwin himself (Figure 1) put it best in the final passage of On the Origin of Species:
APPENDIX
Cranium
Mandible
Cervical vertebrae
Clavicle 1st and 2nd
thoracic vertebrae
Sternum
Mandible Scapula
Scapula
Sternum Humerus
11th and 12th
thoracic vertebrae
Humerus Lumbar
vertebrae
Ilium
Radius
Sacrum
Coccyx
Radius Ulna
Carpus
Metacarpals
Ulna Phalanges
Pubis
Ischium
Femur
Patella
Tibia
Fibula
Metatarsals
Tarsus
Metatarsals
Phalanges
Phalanges Tarsus
A1
This page intentionally left blank
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A3
GLOSSARY
abductor A muscle whose contraction moves a limb away isolated from each other and then diverge to form two
from the midline of the body. The abductors that con- or more new species. Compare parapatric speciation
nect the pelvis to the femur act to keep the body upright and sympatric speciation.
during bipedal walking. altruism (altruistic, adj.) Behavior that reduces the fitness
Acheulean A tool industry found at sites dated at 1.6 to of the individual performing the behavior (the actor)
0.3 mya and associated with Homo ergaster and some but increases the fitness of the individual affected by the
archaic Homo sapiens. Named after the French village behavior (the recipient). Compare mutualism. See also
of Saint-Acheul, where it was first discovered, the selfish and spiteful.
Acheulean industry is dominated by teardrop-shaped amino acids Molecules that are linked in a chain to form
hand axes and blunt cleavers. proteins. There are 20 different amino acids, all of
achondroplasia A genetic disease caused by a dominant which share the same molecular backbone but have a
gene that leads to the development of short stature and different side chain.
disproportionately short arms and legs. aminoacyl-tRNA synthetases Enzymes that link amino
activator A protein that increases transcription of a regu- acids to the appropriate varieties of transfer RNA as
lated gene. Compare repressor. part of protein synthesis.
actor The individual performing a given behavior. Com- analogy (analogous, adj.) Similarity between traits that is
pare recipient. due to convergent evolution, not common descent. For
adaptation A feature of an organism created by the process example, the fact that humans and kangaroos are both
of natural selection. bipedal is an analogy. Compare homology.
adaptive grade The basic way that an animal makes a liv- ancestral trait A trait that appears earlier in the evolution
ing. Distantly related animals can belong to the same of a lineage or clade. Ancestral traits are contrasted
adaptive grade. with derived traits, which appear later in the evolution
adaptive radiation The process in which a single lineage of a lineage or clade. For example, the presence of a tail
diversifies into a number of species, each characterized is ancestral in the primate lineage, and the absence of a
by distinctive adaptations. The diversification of the tail is derived. Systematists must avoid using ancestral
mammals at the beginning of the Cenozoic era is an similarities when constructing phylogenies.
example of an adaptive radiation. angiosperms The flowering plants. The radiation of the
adenine One of the four bases of the DNA molecule. The angiosperms during the Cretaceous period may have
complementary base of adenine is thymine. played an important role in the evolution of the pri-
affiliative Friendly. mates. Compare gymnosperms.
alkaloids Secondary compounds produced and kept in anthropoid Any member of the primate suborder that
plant tissues to make the plant distasteful or even poi- includes the monkeys and apes. The only other primate
sonous to herbivores. suborder (the prosimians) includes lemurs, lorises, and
allele One of two or more alternative forms of a gene. For tarsiers.
example, the A and S alleles are two forms of the gene anticodon The sequence of bases on a transfer RNA mol-
controlling the amino acid sequence of one of the sub- ecule that binds complementarily to a particular codon.
units of hemoglobin. For example, for the codon ATC the corresponding
allele frequency See gene frequency. anticodon is TAG because A binds to T, and G to C.
alliance An interaction in which two or more animals apatite crystal A crystalline material found in tooth
jointly initiate aggression against, or respond to aggres- enamel.
sion from, one or more other animals. Also called arboreal Active predominantly in trees. Compare
coalition. terrestrial.
allopatric speciation Speciation that occurs when two or archaic Homo sapiens An older term for hominins with
more populations of a single species are geographically larger brains and more modern crania that appear in
A3
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A4
A4 Glossary
the fossil record about 500 kya in Africa and Europe, bipedal Describing locomotion in which the animal walks
and somewhat later in eastern Asia. upright on two (hind) legs. Compare quadrupedal.
argon–argon dating A sophisticated variant of the potas- blade A stone tool, made from a flake, that is at least twice
sium–argon dating method that allows very small sam- as long as it is wide. Blades dominate the tool traditions
ples to be dated accurately. of the Upper Paleolithic.
blending inheritance A model of inheritance, widely held
australopithecine Any member of the genus Australop- during the nineteenth century, in which the hereditary
ithecus, consisting of extinct hominins that lived from material of the mother and father was thought to com-
4.2 mya to about 1.8 mya. Australopithecines were bine irreversibly in the offspring.
characterized by bipedal locomotion, robust teeth and brachiation A form of movement in which the body is pro-
jaws, and ape-sized brains. (In this text we recognize six pelled by the arms alone with a phase of free flight
species: A. anamensis, A. afarensis, A. africanus, A. between handholds. Only gibbons and siamangs are
gahri, A. habilis, and A. rudolfensis.) true brachiators.
brainstem The portion of the brain that lies between the
bachelor male A male that has not been able to establish cerebrum and the spinal cord and provides the major
residence in a bisexual group. Bachelor males may live route for communication between the forebrain, the
alone or reside in all-male groups. spinal cord, and the peripheral nerves.
balanced polymorphism A steady state in which two or bridewealth The collection of valuable items that is trans-
more alleles coexist in a population. This state occurs ferred from the groom’s family to the bride’s family at
when heterozygotes have a higher fitness than any the time of marriage.
homozygote.
camera-type eye An eye in which light passes through a
basal metabolic rate The rate of energy use required to transparent opening and is then focused by a lens on
maintain life when an animal is at rest. photosensitive tissue. Camera-type eyes are found in
base One of four molecules—adenine, guanine, cytosine, vertebrates, mollusks, and some arthropods.
and thymine—that are bound to the DNA backbone. canalized Describing traits that are very insensitive to
Different sequences of bases encode the information environmental conditions during development, result-
necessary for protein synthesis. ing in similar phenotypes in a wide range of environ-
basicranium (basicrania, pl.) The base or underside of the ments. Compare plastic.
cranium. canine The sharp, pointed tooth that lies between the
incisors and the premolars in primates.
biface A flat stone tool made by working both sides of a carbohydrates Certain organic molecules with the for-
core until there is an edge along the entire circumfer- mula CnH2nOn, including common sugars and starches.
ence. See also hand ax. carbon-14 dating A dating method based on an unstable
bilaterally symmetrical Describing an animal whose mor- isotope of carbon with atomic weight of 14. Carbon-
phology on one side of the midline is a mirror image of 14 is produced in the atmosphere by cosmic radiation
the morphology on the other side. and is taken up by living organisms. After organisms
binocular vision Vision in which both eyes can focus to- die, the carbon-14 present in their bodies decays to a
gether on a distant object to produce three-dimensional stable isotope (nitrogen-14) at a constant rate. By meas-
images. See also stereoscopic vision. uring the ratio of carbon-14 to the stable isotope of car-
bon (carbon-12) in organic remains, scientists can
biochemical pathway Any of the chains of chemical reac- estimate the length of time that has passed since the
tions by which organisms regulate their structure and organism died. The carbon-14 method is useful for
chemistry. dating specimens that are younger than about 40,000
biological species concept The concept that species are years old. Also called radiocarbon dating.
defined as a group of organisms that cannot interbreed cathemeral Active both during the day and at night.
in nature. Adherents of the biological species concept character A trait or attribute of the phenotype of an
believe that the resulting lack of gene flow is necessary organism.
to maintain differences between closely related species. character displacement The result of competition between
Compare ecological species concept. two species that causes the members of different species
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A5
Glossary A5
to become morphologically or behaviorally more dif- conspecifics Members of the same species.
ferent from each other. contest competition A form of competition that occurs
Châtelperronian An Upper Paleolithic tool industry found when resources are clumped in space and worth defend-
in France and Spain that dates from 36 to 32 kya and ing. Compare scramble competition.
is associated with Neanderthal fossil remains. continental drift The movement over the surface of the
chimera A combination of more than one genetic lineage globe of the immense plates of relatively light material
within a single individual. that make up the continents.
chorion The outer membrane that surrounds the fetus in continuous variation Phenotypic variation in which there
utero and gives rise to the placenta. is a continuum of types. Height in humans is an exam-
chromosome A linear body in the cell nucleus that carries ple of continuous variation. Compare discontinuous
genes and appears during cell division. Staining cells variation.
with dyes reveals that different chromosomes are convergence The evolution of similar adaptations in unre-
marked by different banding patterns. lated species. The evolution of camera-type eyes in
cladistic taxonomy A system for classifying organisms in both vertebrates and mollusks is an example of con-
which patterns of descent are the only criteria used. vergence. See also analogy.
Compare evolutionary taxonomy. core A piece of stone from which smaller flakes are
cleaver A biface stone tool with a broad, flat edge. removed. Cores and/or flakes may themselves be useful
Cleavers are common at Acheulean sites. tools.
coalition See alliance. corpus luteum A mass of cells that forms in a woman’s
coding sequence A DNA sequence that encodes the amino ovary after ovulation and produces the hormone
acid sequence of a protein. progesterone.
codon A sequence of three DNA bases on a DNA molecule correlated characters Traits that are statistically associated
that constitutes one “word” in the message used to cre- in a population. For example, arm length and leg length
ate a specific protein. There are 64 different codons. are correlated if people with long arms also tend to have
Compare anticodon. long legs, and vice versa. Correlated characters arise
coefficient of relatedness (r) An index measuring the degree when particular genes affect multiple characters. See
of genetic closeness between two individuals. The index also pleiotropic effects.
ranges from 0 (for no relation) to 1 (which occurs only correlated response An evolutionary change in one char-
between an individual and itself, or between identical acter caused by selection on a second, correlated char-
twins). For example, the coefficient of relatedness acter. For example, selection favoring only long legs will
between an individual and its parents or its siblings is also increase arm length if arm length and leg length are
0.5. positively correlated.
cognitive map A mental representation of the location of cortex The original, unmodified surface of a stone used to
objects in space and time that allows for efficient make stone tools.
navigation. cross In genetics, a mating between chosen parents.
collected food Type of food resource, such as a leaf or crossing over The exchange of genetic material between
fruit, that can be gathered and eaten directly. Compare homologous chromosomes during meiosis. Crossing
extracted food and hunted food. over causes recombination of genes carried on the same
combinatorial control The control of gene expression in chromosome.
which more than one regulatory protein is used and crural index The ratio of the length of the shin bone
expression is allowed only in a specific combination of (tibia) to the length of the thigh bone (femur).
conditions. CT See computed tomography.
comparative method A method for establishing the func- culture Information stored in human brains that is
tion of a phenotypic trait by comparing different acquired by imitation, teaching, or some other form of
species. social learning and that is capable of affecting behavior
compound eye An eye in which the image is formed by a or some other aspect of the individual’s phenotype.
large number of discrete photoreceptors. Compound cusp A projection on the biting surface of a molar or
eyes are found in insects and other arthropods. premolar.
computed tomography (CT) An X-ray technique that gen- cytoplasm The material that is inside the cell but outside
erates three-dimensional images. the nucleus.
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A6
A6 Glossary
cytosine One of the four bases of the DNA molecule. The dominance hierarchy A ranking of individuals in a group
complementary base of cytosine is guanine. that reflects their relative dominance.
dominance matrix A square table constructed to keep
dental formula The number of incisors, canines, premo- track of dominance interactions among a group of indi-
lars, and molars in the upper and lower jaws. viduals. Usually winners are listed down the left side
deoxyribonucleic acid See DNA. and losers are listed across the top, and the number of
derived trait A trait that appears later in the evolution of times each individual defeats another is entered in the
a lineage or clade. Derived traits are contrasted with cells of the matrix. Individuals are ordered in the matrix
ancestral traits, which appear earlier in the evolution of so as to minimize the number of entries below the
a lineage or clade. For example, the absence of a tail is diagonal. This ordering is then used to construct the
derived in the hominin lineage, and the presence of a dominance hierarchy.
tail is ancestral. Systematists seek to use derived simi- dominant Describing an allele that results in the same phe-
larities when constructing phylogenies. notype whether in the homozygous or the heterozygous
development All of the processes by which the single-celled state. Compare recessive.
zygote is transformed into a multicellular adult. dyadic Describing an interaction that involves two indi-
viduals. Also called pairwise.
diastema (diastemata, pl.) A gap between adjacent teeth.
diploid Referring to cells containing pairs of homologous
ecological species concept The concept that natural selec-
chromosomes, in which one chromosome of each pair
tion plays an important role in maintaining the differ-
is inherited from each parent. Also referring to organ-
ences between species, and that the absence of
isms whose somatic (body) cells are diploid; all primates
interbreeding between two populations is not a neces-
are diploid. Compare haploid.
sary condition for defining them as separate species.
discontinuous variation Phenotypic variation in which
Compare biological species concept.
there are a discrete number of phenotypes with no
EEA See environment of evolutionary adaptedness.
intermediate types. Pea color in Mendel’s experiments
electron-spin-resonance dating A technique used to date
is an example of discontinuous variation. Compare
fossil teeth by measuring the density of electrons
continuous variation.
trapped in apatite crystals in teeth. This method is
discriminative parental solicitude A tendency among par- important for sites that are too young to be dated by
ents to adjust their investment in offspring according to potassium–argon dating (less than 500 kya) and too old
cues that predict (or did predict in past environments) to be dated with carbon-14 dating (more than 40 kya).
the likelihood that the offspring will survive and endocranial volume The volume inside the braincase.
reproduce. environment of evolutionary adaptedness (EEA) The past
diurnal Active only during the day. Compare nocturnal. environment(s) in which currently observed adapta-
See also cathemeral. tions were shaped. For example, the psychological
dizygotic twins Twins that result from the fertilization of mechanisms that cause contemporary humans to
two separate eggs by two separate sperm. Dizygotic overeat were likely shaped in an environment of evolu-
twins are no more closely related than other full sib- tionary adaptedness in which overeating was rarely a
lings. Compare monozygotic twins. problem.
DNA Deoxyribonucleic acid, the molecule that carries environmental covariation The effect on phenotypes that
hereditary information in almost all living organisms. occurs when the environments of parents and offspring
DNA consists of two very long phosphate–sugar back- are similar. Because environmental covariation causes
bones (called “strands”) to which the bases adenine, the phenotypes of parents and offspring to be similar, it
cytosine, guanine, and thymine are bound. Hydrogen can falsely increase estimates of heritability.
bonds between the bases bind the two strands together. environmental variation Phenotypic differences between
dominance The ability of one individual to intimidate or individuals that exist because those individuals devel-
defeat another individual in a pairwise (dyadic) en- oped in different environments. Compare genetic
counter. In some cases, dominance is assessed from the variation.
outcome of aggressive encounters; in other cases, dom- enzyme A protein that serves as a catalyst, increasing the
inance is assessed from the outcome of competitive rate at which particular chemical reactions occur at a
encounters. given temperature. Enzymes can control the chemical
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A7
Glossary A7
composition of cells by causing some chemical reactions food sharing The practice of sharing food within a social
to occur much faster than others. group. Food sharing has been widely observed among
equilibrium A steady state in which neither gene frequen- contemporary human foragers but is relatively uncom-
cies nor genotypic frequencies change. mon in other primate species.
estrus A period during the reproductive cycle of most foramen magnum The large hole in the bottom of the cra-
mammals (and most primates) when the female is nium through which the spinal cord passes.
receptive to mating and is capable of conceiving. fossil A trace of life more than 10,000 years old preserved
eukaryotes Organisms whose cells have cellular organelles, in rock. Fossils can be mineralized bones, plant parts,
cell nuclei, and chromosomes. All plants and animals impressions of soft body parts, or tracks.
are eukaryotes. Compare prokaryotes. founder effect A form of genetic drift that occurs when a
evolutionary taxonomy A system for classifying organisms small population colonizes a new habitat and subse-
that uses both patterns of descent and patterns of over- quently greatly increases in number. Random genetic
all similarity. Compare cladistic taxonomy. changes due to the small size of the initial population
executive brain Composed of the neocortex and the stria- are amplified by subsequent population growth.
tum, a structure in the basal ganglia that is functionally fraternal polyandry A marriage system in which two or
linked to the neocortex. more brothers marry a single wife. Fraternal polyandry
exon A segment of the DNA in eukaryotes that is trans- is practiced among the Nyinba, a Tibetan-speaking
lated into protein. Compare intron. group living in western Nepal.
extracted food Food that is embedded in a matrix, encased frugivore (frugivorous, adj.) An animal whose diet consists
in a hard shell, or otherwise difficult to extract. mostly of fruit.
Extracted foods require complicated, carefully coordi-
nated techniques to process. Compare collected food
and hunted food. gametes In animals, eggs and sperm.
gene A segment of the chromosome that produces a rec-
ognizable effect on phenotype and segregates as a unit
F0, F1, and F2 generations A system for keeping track of during gamete formation.
generations in breeding experiments. The initial gener- gene flow The movement of genes from one population to
ation is called the F0 generation, the offspring of the F0 another, or from one part of a population to another, as
generation constitute the F1 generation, and the off- the result of interbreeding.
spring of the F1 generation constitute the F2 generation. gene frequency The fraction of the genes at a genetic locus
falciparum malaria A severe form of malaria. The sickle- that are a particular allele (therefore also called allele
cell allele for hemoglobin is common in West Africa frequency). For example, a population that contains
because it confers resistance to falciparum malaria in 250 AA individuals, 200 AS individuals, and 50 SS indi-
the heterozygous state. viduals has 700 copies of the A allele and 300 copies of
family A taxonomic level above genus but below order. A the S allele; therefore the frequency of the S allele is 0.3.
family may contain several genera, and an order may gene tree A phylogenetic tree tracing the pattern of descent
contain several families. Humans belong to the family for a particular gene.
Hominidae, and the other great apes belong to the genetic distance A measure of the overall genetic similar-
family Pongidae. ity of individuals or species. The best estimates of
fecundity The biological capacity to reproduce. In humans, genetic distance utilize large numbers of genes.
fecundity may be greater than fertility (the actual num- genetic drift Random change in gene frequencies due to
ber of children produced) when people limit family size. sampling variation that occurs in any finite population.
femur The thigh bone. Genetic drift is more rapid in small populations than in
fixation A state that occurs when all of the individuals in large populations.
a population are homozygous for the same allele at a genetic variation Phenotypic differences between individ-
particular locus. uals that result from the fact that those individuals have
flake A smaller chip of stone knocked from a larger stone inherited different genes from their parents. Compare
core. environmental variation.
folivore (folivorous, adj.) An animal whose diet consists genome All of the genetic information carried by an
mostly of leaves. organism.
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A8
A8 Glossary
genotype The combination of alleles that characterizes an haplorhine Any member of the group containing tarsiers
individual at some set of genetic loci. For example, in and anthropoid primates. The system that classifies
populations with only the A and S alleles at the hemo- primates into haplorhines and strepsirhines is a cladis-
globin locus, that locus has only three possible geno- tic alternative to the system used in this text, in which
types: AA, AS, and SS. (SA is the same as AS.) Compare primates are divided into prosimians and anthropoids,
phenotype. and tarsiers are grouped with prosimians. Compare
genotypic frequency The fraction of individuals in a pop- strepsirhine.
ulation that have a particular genotype. haplotype A particular set of alleles at some number of
genus (genera, pl.) A taxonomic category below family and genetic loci that are transmitted together on the same
above species. There may be several species in a genus, chromosome.
and several genera in a family. Hardy–Weinberg equilibrium The unchanging frequency
Gondwanaland The more southerly of the two supercon- of genotypes that results from sexual reproduction and
tinents that existed from about 180 mya to roughly 140 occurs in the absence of other evolutionary forces such
mya. Gondwanaland included the continental plates as natural selection, mutation, or genetic drift.
that now make up Africa, South America, Antarctica, hCG See human chorionic gonadotropin.
Australia, New Guinea, Madagascar, and the Indian hemoglobin A protein in blood that carries oxygen, includ-
subcontinent. Compare Laurasia. ing two α (alpha) and two β (beta) subunits.
grammar All of the rules of meaning that our brains use to heritability The fraction of the phenotypic variation in the
interpret information provided by language. population that is the result of genetic variation.
grooming The process of picking through hair to remove heterozygous Referring to a diploid organism whose cells
dirt, dead skin, ectoparasites, and other material. carry two different alleles for a particular genetic locus.
Grooming is a common form of affiliative behavior Organisms that are heterozygous are called “heterozy-
among primates. gotes.” Compare homozygous.
guanine One of the four bases of the DNA molecule. The hindlimb-dominated Describing an animal that habitually
complementary base of guanine is cytosine. carries most of its weight on its hindlimbs.
gum A sticky carbohydrate produced by some trees in home base A temporary camp that members of a group
response to physical damage. Gum is an important return to each day. At the home base, food is shared,
food for many primates. processed, cooked, and eaten; subsistence tools are
gummivore (gummivorous, adj.) An animal whose diet manufactured and repaired; and social life is conducted.
consists mostly of gum. home range The area in which an individual or a group of
gymnosperms A group of plants that reproduce without animals travels, feeds, rests, and socializes. Territorial
flowering. Modern gymnosperms include pines, red- species actively defend the borders of their home ranges.
woods, and firs. Compare angiosperms. hominin Any member of the family Hominidae, including
all species of Australopithecus and Homo.
haft To attach a spear point, ax head, or similar implement hominoid Any member of the superfamily Hominoidea,
to a handle. Hafting greatly increases the force that can which includes humans, all the living apes, and numer-
be applied to the tool. ous extinct apelike and humanlike species from the
Hamilton’s rule A rule predicting that altruistic behavior Miocene, Pliocene, and Pleistocene epochs.
among relatives will be favored by natural selection if Homo heidelbergensis Middle Pleistocene hominins from
rb > c, where r is the coefficient of relatedness between Africa and western Eurasia. These hominins had large
actor and recipient, b is the sum of the benefits of per- brains and very robust skulls and postcrania.
forming the behavior on the fitness of the recipient(s), homologous chromosomes A pair of chromosomes in a
and c is the cost, in decreased fitness of the donor, of diploid cell in which one member of the pair is derived
performing the behavior. See also kin selection. from the father and one is derived from the mother.
hand ax The most common type of biface stone tool homologous pair See homologous chromosomes.
found in Acheulean sites. It is flat and teardrop-shaped, homology (homologous, adj.) Similarity between traits
with a sharp point at the narrow end. that is due to common ancestry, not convergence. For
haploid A cell with only one copy of each chromosome. example, the reason that gorillas and baboons are both
Gametes are haploid, as are the cells of some asexual quadrupedal is that they are both descended from a
organisms. Compare diploid. quadrupedal ancestor. Compare analogy.
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A9
Glossary A9
homozygous Referring to a diploid organism whose chro- such competition, like large body size or large canines,
mosomes carry two copies of the same allele at a single are selected for. Compare intersexual selection.
genetic locus. Organisms that are homozygous are intron A segment of the DNA in eukaryotes that is not
called “homozygotes.” Compare heterozygous. translated into protein. Compare exon.
human chorionic gonadotropin (hCG) A hormone, isotope A chemical element with the same atomic number
secreted by the embryo during the early part of preg- and properties as another element but having a differ-
nancy, that stimulates the mother’s body to produce ent atomic weight. Unstable isotopes spontaneously
progesterone. change into more stable isotopes. See also radioactive
humerus The bone in the upper part of the forelimb (arm). decay.
hunted food Live animal prey captured by human foragers
or nonhuman primates. Compare collected food and kibbutz (kibbutzim, pl.) An agricultural settlement in
extracted food. Israel, usually organized according to collectivist
hybrid zone A geographic region where two or more pop- principles.
ulations of the same species or two different species kin selection A theory stating that altruistic acts will be
overlap and interbreed. Hybrid zones usually occur at favored by selection if the product of the benefit to the
the habitat margins of the respective populations. recipient and the degree of relatedness (r) between the
actor and recipient exceeds the cost to the actor. See
ilium (ilia, pl.) One of the three bones in the pelvis. also Hamilton’s rule.
inbred mating Mating between closely related individuals. knapping The process of manufacturing stone tools.
Also called inbreeding. Compare outbred mating.
knuckle walking A form of quadrupedal locomotion in
inbreeding See inbred mating.
which, in the forelimbs, weight is supported by the
incisors The front teeth in mammals. In anthropoid pri-
knuckles, rather than by the palm or outstretched fin-
mates, incisors are used for cutting, and there are two
gers. Chimpanzees and gorillas are knuckle walkers.
on each side of the upper and lower jaw.
independent assortment The principle, discovered by
lactase persistence Retention of the capacity to synthesize
Mendel, that each of the genes at a single locus on a
the enzyme lactase, which is necessary to digest the
pair of homologous chromosomes is equally likely to be
main carbohydrates in fresh milk after weaning.
transmitted when gametes (eggs and sperm) are formed.
lactase-phlorizin hydrolase (LHP) An enzyme produced in
This happens because during meiosis the probability
the small intestine that breaks down lactose in milk.
that a particular chromosome will enter a gamete is 0.5
The term is frequently shortened to lactase.
and is independent of whether other nonhomologous
chromosomes enter the same gamete. Thus, knowing lactation (lactate, v.) Production of milk by the mammary
that an individual received a particular chromosome glands in females; also, the period during which milk is
from its mother (and thus a particular allele) tells you produced for nursing offspring. Lactation is a charac-
nothing about the probability that it received other, teristic feature of mammals.
nonhomologous chromosomes from its mother. lactose A sugar present in mammalian milk. Most
infraorder The taxonomic level between order and super- mammals—including most humans—lose the ability to
family. An order may contain several infraorders, and digest lactose as adults.
an infraorder may contain several superfamilies. Later Stone Age (LSA) The tool industries found in Africa
insectivore (insectivorous, adj.) An animal whose diet con- that correspond in age and type to the Upper Paleolithic
sists mostly of insects. industries in Europe.
insulin A substance that is created by the pancreas and is Laurasia The more northerly of the two supercontinents
involved in the regulation of blood sugar. that existed from roughly 180 mya to 140 mya. Laura-
intersexual selection A form of sexual selection in which sia included what is now North America, Greenland,
females choose who they mate with. The result is that Europe, and parts of Asia. Compare Gondwanaland.
traits making males more attractive to females are LCT The structural gene that codes for lactase-phlorizin
selected for. Compare intrasexual selection. hydrolase (lactase).
intrasexual selection A form of sexual selection in which Levallois technique A three-step toolmaking method used
males compete with other males for access to females. by Neanderthals. The knapper first makes a core hav-
The result is that traits making males more successful in ing a precisely shaped convex surface, then makes a
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A10
A10 Glossary
striking platform at one end of the core, and finally gene expression. Some are involved in regulating devel-
knocks a flake off the striking platform. opment and cell differentiation in complex organisms.
LH See luteinizing hormone. microsatellite loci Regions within DNA sequences in which
linked Referring to genes located on the same chromo- short sequences of DNA are repeated multiple times—
some. The closer together two loci are, the more likely GTGTGT or ACTACTACT. They are also known as
they are to be linked. Compare unlinked. short tandem repeats.
locus (loci, pl.) The position on a chromosome that is Middle Stone Age (MSA) The stone tool industries of sub-
occupied by a particular gene. Saharan Africa and southern and eastern Asia that
LSA See Later Stone Age. existed 250 to 40 kya. The MSA is the counterpart of
luteinizing hormone (LH) A hormone secreted by the pitu- the Middle Paleolithic (Mousterian) in Europe. The
itary gland. In females, LH stimulates ovulation, for- MSA industries varied, but flake tools were manufac-
mation and maintenance of the corpus luteum, and tured in all of them.
production of the hormone progesterone. In males, LH mineralization (mineralized, adj.) The process by which
stimulates production of the hormone androgen. organic material in the bones of dead animals is
replaced by minerals from the surrounding rock, creat-
macroevolution Evolution of new species, families, and ing fossils.
higher taxa. Compare microevolution. minor marriage A form of marriage, formerly widespread
maladaptive Detrimental to fitness. in China, in which children were betrothed in infancy
mandible The lower jaw. Compare maxilla. and then raised together in the household of the
marsupial A mammal that gives birth to live young that prospective groom.
continue their development in a pouch equipped with mitochondrial DNA (mtDNA) DNA in the mitochondria
mammary glands. Marsupials include kangaroos and that is particularly useful for evolutionary analyses, for
opossums. two reasons: (1) mitochondria are inherited only from
mate guarding A form of mating in which the male defends the mother, and thus there is no recombination, and (2)
his mate after copulation to prevent other males from mtDNA accumulates mutations at relatively high rates,
mating with her. thus serving as a more accurate molecular clock for
mating efforts Efforts made to secure access to a mate, recent changes (last few million years).
such as courtship displays or provision of resources to mitochondrion (mitochondria, pl.) A cellular organelle
the prospective mate. that is involved in basic energy processing.
mating system The form of courtship, mating, and par- mitosis The process of division of somatic (normal body)
enting behavior that characterizes a particular species cells through which new diploid cells are created. Com-
or population. An example is polygyny. pare meiosis.
matrilineage Individuals related through the maternal line. Mode 1 A category of simple stone tools made by remov-
maxilla The upper jaw. Compare mandible. ing flakes from cores without any systematic shaping of
meiosis The process of cell division in which haploid the core. Both the flakes and the cores were probably
gametes (eggs and sperm) are created. Compare mitosis. used as tools themselves. Tools in the Oldowan indus-
meme A term coined by Richard Dawkins to refer to a unit try are Mode 1 tools.
of cultural information (belief or value) transmitted by Mode 2 A category of stone tools in which cores are
imitation and teaching. shaped into symmetrical bifaces by the removal of
menarche First menstruation. flakes. The Acheulean industry is typified by Mode 2
messenger RNA (mRNA) A form of RNA that carries tools.
specifications for protein synthesis from DNA to the Mode 3 A category of stone tools made by striking large
ribosomes. symmetrical flakes from carefully prepared stone cores
microevolution Evolution of populations within a species. using the Levallois technique. The Mousterian industry
Compare macroevolution. in Europe and the Middle Stone Age industries in Africa
microlith A very small stone flake. Typical of African are typified by Mode 3 tools.
Later Stone Age industries, microliths were probably Mode 4 A category of stone tools in which blades are com-
hafted onto wood handles to make spears and axes. mon. Mode 4 tools are found in some Middle Stone
microRNA (miRNA) Short segments of RNA that are Age industries in Africa, and they predominate in the
involved in the translation of mRNA into protein and Upper Paleolithic industries of Europe.
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A11
Glossary A11
Mode 5 A category of stone tools in which microliths are Neanderthal A form of archaic Homo sapiens found in
common. The African Later Stone Age industries are western Eurasia from about 127 kya to about 30 kya.
typified by Mode 5 tools. Neanderthals had large brains and elongated skulls
modern synthesis An explanation for the evolution of with very large faces. They were also characterized by
continuously varying traits that combines the theory very robust bodies.
and empirical evidence of both Mendelian genetics and negative selection Selection against novel mutants that
Darwinism. preserves the existing genotype.
molars The broad, square back teeth that are generally negatively correlated Describing a statistical relationship
adapted for crushing and grinding in primates. Anthro- between two variables in which larger values of one
poid primates have three molars on each side of the variable tend to co-occur with smaller values of the
upper and lower jaw. other variable. For example, the size and number of
molecular clock The hypothesis that genetic change occurs seeds produced by an individual plant are negatively
at a constant rate and thus can be used to measure the correlated in some plant populations. Compare posi-
time elapsed since two species shared a common ances- tively correlated.
tor. The molecular clock is based on observed regular- neocortex Part of the cerebral cortex; generally thought to
ities in the rate of genetic change along different be most closely associated with problem solving and
phylogenetic lines. behavioral flexibility. In mammals, the neocortex cov-
monozygotic twins Twins that result from the fertilization ers virtually the entire surface of the forebrain.
of one egg by a single sperm. Early in development the neocortex ratio The size of the neocortex in relation to the
fertilized egg splits to create two zygotes. Compare rest of the brain.
dizygotic twins. nest parasitism A behavior seen in some bird species in
morphology The form and structure of an organism; also which females lay eggs in the nests of other birds,
a field of study that focuses on the form and structure which then unwittingly raise the alien chicks as their
of organisms. own.
Mousterian A stone tool industry characterized by points, neutral theory A theory postulating that genetic change is
side scrapers, and denticulates (tools with small tooth- caused only by mutation and drift.
like notches on the working edge), but an absence of niche The way of life, or “trade,” of a particular species—
hand axes. The Mousterian is generally associated with what foods it eats and how the food is acquired.
Neanderthals in Europe. NIDD See non-insulin-dependent diabetes.
mRNA See messenger RNA. nocturnal Active only during the night. Compare diurnal.
MSA See Middle Stone Age. See also cathemeral.
mtDNA See mitochondrial DNA. noncoding RNA Molecules of RNA that do not code for
mutation A spontaneous change in the chemical structure proteins, including transfer RNA, ribosomal RNA, and
of DNA. microRNAs.
mutualism (mutualistic, adj.) Behavior that increases the non-insulin-dependent diabetes (NIDD) A form of diabetes
fitness of both actor and recipient. Compare altruism. in which cells of the body do not respond properly to
See also selfish and spiteful. levels of insulin in the blood. NIDD is known to have
a genetic basis.
natal group The group into which an individual is born. In nonsynonymous substitution Substitution of one
many primate species the females remain in their natal nucleotide for another in a DNA sequence that changes
groups throughout their lives, while the males emi- the amino acid coded for. Compare synonymous
grate and join new groups. substitution.
natural selection The process that produces adaptation. nucleus (nuclei, pl.) The distinct part of the cell that con-
Natural selection is based on three postulates: (1) the tains the chromosomes. Eukaryotes (fungi, protozoans,
availability of resources is limited; (2) organisms vary in plants, and animals) all have nucleated cells; prokary-
the ability to survive and reproduce; and (3) traits that otes (bacteria) do not.
influence survival and reproduction are transmitted
from parents to offspring. When these three postulates observational learning A form of learning in which ani-
hold, natural selection produces adaptation. mals observe the behavior of other individuals and
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A12
A12 Glossary
thereby learn to perform a new behavior. Compare phenotype The observable characteristics of organisms.
social facilitation. Individuals with the same phenotype may have different
occipital torus A horizontal ridge at the back of the skull genotypes. Compare genotype.
in Homo erectus and archaic Homo sapiens. phenotypic matching A mechanism for kin recognition in
Oceania A region of the South Pacific that includes Poly- which animals assess similarities between themselves
nesia, Melanesia, and Micronesia. and others.
philopatry The tendency in some animals to remain in
Oldowan A set of simple stone tools made by removing
their natal (birth) groups throughout their lives. In
flakes from cores without any systematic shaping of the
many Old World monkey species, females are philo-
core. Both the flakes and the cores were probably used
patric.
as tools. This industry is found in Africa at sites that
phoneme The basic unit of speech perception. Phonemes
date from about 2.5 mya.
are the smallest bits of sound that we recognize as
olfaction (olfactory, adj.) The sense of smell.
meaningful elements of language. People who speak dif-
opposable Describing the property of the thumb or big toe ferent languages recognize slightly different sets of
that enables some primates (Old World monkeys and phonemes.
apes) to touch the tips of the thumb and forefinger phylogeny The evolutionary relationships among a group
together. of species, usually diagrammed as a “family tree.”
organelle A portion of the cell that is enclosed in a mem- pick A triangular-shaped biface stone tool found in
brane and has a specific function; examples are mito- Acheulean sites.
chondria and the nucleus. placental mammal A mammal that gives birth to live
out-group A taxonomic group that is related to a group of young that developed for a period of time in the uterus
interest and can be used to determine which traits are and were nourished by blood delivered to a placenta.
ancestral and which are derived. plastic Describing traits that are very sensitive to environ-
outbred mating Mating between unrelated individuals. mental conditions during development, resulting in dif-
Compare inbred mating. ferent phenotypes in different environments. Compare
canalized.
pair-bonded A form of social organization in which pleiotropic effects Phenotypic effects created by genes that
groups are composed of a single mated pair and their influence multiple characters. See also correlated
dependent offspring. characters.
pairwise Describing an interaction that involves two indi- plesiadapiform Any member of a group of primatelike
viduals. Also called dyadic. mammals that lived during the Paleocene (65 to 55
mya). Although many paleontologists do not consider
paleontologist A scientist who studies fossilized remains of
them to have been primates, the plesiadapiforms prob-
plant and animal species.
ably were similar to the earliest primates, who lived
Pangaea The massive single continent that contained all of
around the same time.
the Earth’s dry land until about 125 mya.
pneumatized Containing cavities filled with air.
parapatric speciation A two-step process of speciation in polyandry A mating system in which a single female forms
which (1) selection causes the differentiation of geo- a stable pair-bond with two different males at the same
graphically separate, partially isolated populations of a time. Polyandry is generally rare among mammals, but
species and (2) subsequently the populations become it is thought to occur in some species of marmosets and
reproductively isolated as a result of reinforcement. tamarins. Compare polygyny.
Compare allopatric speciation and sympatric specia- polygyny A mating system in which a single male mates
tion. with many females. Polygyny is the most common mat-
parent–offspring conflict Conflict that arises between par- ing system among primate species. Compare polyandry.
ents and their offspring over how much the parents will population genetics The branch of biology dealing with the
invest in the offspring. These conflicts stem from the processes that change the genetic composition of pop-
opposing genetic interests of parents and offspring. ulations through time.
parenting efforts Efforts made to enhance the survivorship porphyria variegata A genetic disease caused by a domi-
of offspring, such as lactation and infant care. nant gene in which carriers of the gene develop a severe
pastoralists People who make a living herding livestock. reaction to certain anesthetics.
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A13
Glossary A13
positive selection Selection that favors novel genotypes and at a constant rate that can be measured precisely in the
thus leads to genetic change. laboratory.
positively correlated Describing a statistical relationship radiocarbon dating See carbon-14 dating.
between two variables in which larger values of one radiometric method Any dating method that takes advan-
variable tend to co-occur with larger values of the other tage of the fact that isotopes of certain elements change
variable. For example, the height and weight of indi- spontaneously from one isotope to another at a con-
viduals are positively correlated in human populations. stant rate.
Compare negatively correlated. rain shadow An area of reduced rainfall found on the lee
postcranium (postcrania, pl.; postcranial, adj.) The skele- (downwind) side of large mountains and mountain
ton excluding the skull. ranges.
potassium–argon dating A radiometric method of dating recessive Describing an allele that is expressed in the phe-
the age of a rock or mineral by measuring the rate at notype only when it is in the homozygous state. Com-
which potassium-40, an unstable isotope of potassium, pare dominant.
is transformed into argon. This method can be used to recipient The individual to whom a particular behavior is
date volcanic rocks that are at least 500,000 years old. directed. Compare actor.
prehensile Describing the ability of hands, feet, or tails to reciprocal altruism A theory that altruism can evolve if
grasp objects, such as food items or branches. pairs of individuals take turns giving and receiving
premolars The teeth that lie between the canines and altruism over the course of many encounters.
molars. recombination The creation of novel genotypes as a result
primary structure The sequence of amino acids that make of the random segregation of chromosomes and of
up a protein. crossing over.
redirected aggression A behavior in which the recipient of
primiparous Refers to a female who has given birth for the
aggression threatens or attacks a previously uninvolved
first time.
party. For instance, if A attacks B and B then attacks C,
proconsulid Any member of a group of early-Miocene
B’s attacks are an example of redirected aggression.
hominoids that includes the genus Proconsul.
regulatory gene A DNA sequence that regulates the expres-
progesterone An ovarian steroid that plays an important
sion of a structural gene often by binding to an activa-
role in preparing the fetus to sustain a pregnancy.
tor or repressor.
prognathic See subnasal prognathism.
regulatory sequence A regulatory gene.
prokaryotes Organisms that lack a cell nucleus or separate reinforcement The process in which selection acts against
chromosomes. Bacteria are prokaryotes. Compare the likelihood of hybrids occurring between members of
eukaryotes. two phenotypically distinctive populations, leading to
prosimian Any member of the primate suborder that con- the evolution of mechanisms that prevent interbreeding.
tains the lemurs, lorises, and tarsiers. Compare repressor A protein that decreases transcription of a regu-
anthropoid. lated gene. Compare activator.
protein A large molecule that consists of a long chain of reproductive isolation A relationship between two popu-
amino acids. Many proteins are enzyme catalysts; oth- lations in which there is no gene flow between them.
ers perform structural functions. reproductive skew The extent to which reproduction is
protein coding genes Genes that encode instructions for evenly distributed among individuals of one sex within
making proteins. a group. If reproductive skew is high, reproduction is
Punnett square A diagram that uses gene (or allele) fre- monopolized by a small number of individuals. If it is
quencies to calculate the genotypic frequencies for the low, reproduction is egalitarian.
next generation. ribonucleic acid See RNA.
ribosome A small organelle composed of protein and
quadrupedal Describing locomotion in which the animal nucleic acid that temporarily holds together the mes-
moves on all four limbs. Compare bipedal. senger RNA and transfer RNAs during protein
synthesis.
radioactive decay Spontaneous change from one isotope of RNA Ribonucleic acid, a long molecule that plays several
an element to another isotope of the same element or to different important roles in protein synthesis. RNA dif-
an entirely different element. Radioactive decay occurs fers from DNA in that it has a slightly different chemi-
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A14
A14 Glossary
cal backbone and it contains the base thymine instead silverback A mature male gorilla. The term derives from
of uracil. the fact that the hair on the back and shoulders turns
rock shelter A site sheltered by an overhang of rock. silvery gray in mature males.
single nucleotide polymorphism (SNP) A site in the
sagittal crest A sharp fin of bone that runs along the mid- genome where DNA varies from one person to the next
line of the skull that increases the area available for the by a single nucleotide.
attachment of chewing muscles. SLI See specific language impairment.
sagittal keel A feature running along the midline of the SNP See single nucleotide polymorphism.
skull shaped like a shallow, upside-down V. The sagit- social facilitation The situation that occurs when the per-
tal keel is a derived characteristic of Homo erectus. formance of a behavior by older individuals increases
sampling variation The variation in the composition of the probability that younger individuals will acquire
small samples drawn from a large population. that behavior on their own. Social facilitation does not
scapula (scapulae, pl.) Shoulder blade. mean that young individuals copy the behavior of older
scramble competition A form of competition that occurs individuals. For example, the feeding behavior of older
when resources are distributed evenly through space individuals may bring younger individuals in contact
and not worth defending. Compare contest with the foods that adults are eating and, therefore,
competition. increase the chance that they acquire a preference for
secondary compounds Toxic (poisonous) chemical com- those foods. Compare observational learning.
pounds produced by plants and concentrated in plant social intelligence hypothesis The hypothesis that the rel-
tissues to prevent animals from eating the plant. atively sophisticated cognitive abilities of higher pri-
selection–mutation balance An equilibrium that occurs mates are the outcome of selective pressures that
when the rate at which selection removes a deleterious favored intelligence as a means to gain advantages in
gene is balanced by the rate at which mutation intro- social groups.
duces that gene. The frequency of genes at selection– socioecology The study of how social structure is influ-
mutation balance is typically quite low. enced by ecological conditions. Socioecological models
selective sweep A process in which one allele increases in posit that the distribution of resources influences com-
a population due to positive selection. petitive regimes, which in turn influence the distribution
selfish Describing behavior that increases the fitness of the of females, dispersal patterns, the nature of dominance
actor and decreases the fitness of the recipient. Com- relationships, and the quality of social bonds among
pare spiteful. See also altruism and mutualism. females.
sex ratio The number of individuals of one sex in relation solitary A term used for animals that do not live in social
to the number of the opposite sex. By convention, sex groups and do not form regular associations with con-
ratios are generally expressed as the number of males to specifics.
the number of females. species (sing. and pl.) A group of organisms classified
sexual dimorphism Differences between sexually mature together at the lowest level of the taxonomic hierarchy.
males and females in body size or morphology. Biologists disagree about how to define a species (see
sexual selection A form of natural selection that results biological species concept and ecological species con-
from differential mating success in one gender. In mam- cept).
mals, sexual selection usually occurs in males and may specific language impairment (SLI) A family of language
be due to male–male competition. disorders in which the affected person experiences dif-
sexual selection infanticide hypothesis A hypothesis pos- ficulty using language but is of otherwise normal intel-
tulating that infanticide has been favored by sexual ligence. Evidence suggests that at least some cases of SLI
selection because males who kill unweaned infants are are hereditary.
able to enhance their own reproductive prospects if they spiteful Describing behavior that is costly both to the
(1) kill infants whose deaths hasten their mothers’ actor and to the recipient. Compare selfish. See also
resumption of cycling, (2) do not kill their own infants, altruism and mutualism.
and (3) are able to mate with the mothers of the infants splicesomes Organelles that splice the mRNA in eukary-
that they kill. otes after the introns have been snipped out.
sickle-cell anemia A severe form of anemia that afflicts stabilizing selection Selection pressures that favor average
people who are homozygous for the sickle-cell gene. phenotypes. Stabilizing selection reduces the amount of
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A15
Glossary A15
variation in the population but does not alter the mean tapetum (tapeta, pl.) A layer behind the retina that reflects
value of the trait. light in some organisms.
stereoscopic vision Vision in which three-dimensional taphonomy The study of the processes that affect the state
images are produced because each eye sends a signal of of the remains of an organism from the time the organ-
the visual image to both hemispheres in the brain. ism dies until it is fossilized.
Stereoscopic vision requires binocular vision. taurodont root A single broad tooth root in molars, result-
strategy A complex of behaviors deployed in a specific ing from the fusion of three roots. Taurodont roots
functional context, such as mating, parenting, or were characteristic of Neanderthals.
foraging. taxonomy A branch of biology that is concerned with the
stratum (strata, pl.) A geological layer. use of phylogenies for naming and classifying organ-
isms. Compare systematics.
strepsirhine Any member of the group containing lemurs
and lorises. The system classifying primates into hap- temporalis muscle A large muscle involved in chewing. The
lorhines and strepsirhines is a cladistic alternative to the temporalis muscles attach to the side of the cranium
system used in this text, in which primates are divided and to the mandible.
into prosimians and anthropoids, and tarsiers are terrestrial Active predominantly on the ground. Compare
grouped with prosimians. Compare haplorhine. arboreal.
striatum A structure composed of two of the basal ganglia territory A fixed area occupied by animals that defend the
of the forebrain: the caudate nucleus and the putamen. boundaries against intrusion by other individuals or
groups of the same species.
structural gene A DNA sequence that encodes the amino
tertiary structure The three-dimensional folded shape of a
acid sequence of a protein.
protein.
subnasal prognathism The condition in which the part of
testes (testis, sing.) The male organs responsible for sperm
the face below the nose is pushed out.
production.
superfamily The taxonomic level that lies between infra- theory of mind The capacity to be aware of the thoughts,
order and family. An infraorder may contain several knowledge, or perceptions of other individuals. A the-
superfamilies, and a superfamily may contain several ory of mind may be a prerequisite for deception, imita-
families. For example, humans are a member of the tion, teaching, and empathy. It is generally thought that
superfamily Hominoidea, which contains the families humans, and possibly chimpanzees, are the only pri-
Hominidae and Pongidae. mates to possess a theory of mind.
sutures Wavy joints between bones that mesh together and thermoluminescence dating A technique used to date crys-
are separated by fibrous tissue. talline materials by measuring the density of trapped
sympatric speciation A hypothesis that speciation can electrons in the crystal lattice. This method is important
result from selective pressures favoring different phe- for sites that are too young to be dated using potas-
notypes within a population, without positing geo- sium–argon dating (less than 500,000 years old) and
graphic isolation as a factor. Compare allopatric too old to be dated with carbon-14 dating (more than
speciation and parapatric speciation. 40,000 years old).
synonymous substitution Substitution of one nucleotide third-party relationships Relationships among other indi-
for another in a DNA sequence that does not change viduals. For example, monkeys and apes are believed to
the amino acid coded for. Compare nonsynonymous understand something about the nature of kinship rela-
substitution. tionships among other group members.
syntax The grammatical rules that allow people to assign thymine One of the four bases of the DNA molecule. The
meaning to strings of words. complementary base of thymine is adenine.
systematics A branch of biology that is concerned with the tibia The larger of the two long bones in the lower leg.
procedures for constructing phylogenies. Compare torque A twisting force that generates rotary motion.
taxonomy. toxin A chemical compound that is poisonous or toxic.
trait A characteristic of an organism.
tactical deception The use of normal parts of an animal’s transfer RNA (tRNA) A form of RNA that facilitates pro-
behavioral repertoire in an unusual context to achieve tein synthesis by first binding to amino acids in the cyto-
specific objectives that are beneficial to the actor. plasm and then binding to the appropriate site on the
18_Appendix & Glossary [2p]:How Humans Evolved [5e] 10/30/08 12:27 PM Page A16
A16 Glossary
mRNA molecule. There is at least one distinct form of uracil One of the four bases of the RNA molecule. Uracil
tRNA for each amino acid. corresponds to the base thymine in DNA; as with
transitive Describing a property of triadic (three-way) thymine, its complementary base is adenine.
relationships in which the relationships between the
first and second elements and the second and third ele- variant The particular form of a trait. For example, blue
ments automatically determine the relationship between eyes, brown eyes, and gray eyes are variants of the trait
the first and third elements. For example, if A is greater eye color.
than B and B is greater than C, then A is greater than variation among groups Differences in the average phe-
C. In many primate species, dominance relationships notype or genotype between groups.
are transitive. variation within groups Differences in phenotype or geno-
type between individuals in a group.
transposable elements Segments of DNA that move from
vertical clinging and leaping A form of locomotion in
one location to another within the genome of a single
which the animal clings to vertical supports and moves
individual.
by leaping from one vertical support to another.
tRNA See transfer RNA.
vestibular system A system of tubes that are embedded in
the inner ear (inside the cranium) and are part of the
unlinked Referring to genes on different chromosomes. system that animals use to maintain balance.
Compare linked. viviparity Giving birth to live young.
Upper Paleolithic The period from about 45 kya to about
10 kya in Europe, North Africa, and parts of Asia. The zygomatic arch A cheekbone.
tool kits from this period are dominated by blades. zygote The cell formed by the union of an egg and a sperm.
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A17
CREDITS
A17
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A18
A18 Credits
Source/Photo Researchers, Inc; 4.30: From Figure 8 in Sibley of Japanese monkeys, Primates 20:371–397; 6.8a: From Table
and Alquist, 1987, Journal of Molecular Evolution 26:99– 6.1 in S.C. Alberts and J. Altman, 2003, Matrix models for pri-
121. mate life history analysis, In: P. Kappeler and M. E. Pereira,
(eds.). Primate Life History and Socioecology, pp 66–102, Uni-
versity of Chicago Press: Chicago; 6.8b: From 3a in A. M. Rob-
Part II: Baboon footprint (Joan B. Silk) bins, M. R. Robbins, N. Gerald-Steklis, and H.D. Steklis,
2006, Age-related patterns of reproductive success among
female mountain gorillas, Amer. J. Phys. Anthropol. 131: 511–
CHAPTER 5::
521 © Wiley-Liss, Inc., A Wiley Company. Reprinted with per-
5.0: Photograph by John Haskew, Mondika Research Center; mission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons,
Courtesy of Diane Doran; 5.1a: ardea.com/Liz Bomford; 5.1b: Inc.; 6.10b: © Bazuki Muhammad/Reuters/Corbis; 6.11: Cour-
Grospas/Nature; 5.1c: I. Polunin/NHPA; 5.1d: Jan Lind- tesy of Carola Borries; 6.13a-b: From Figures 1 and 2 in A.
blad/Photo Researchers, Inc; 5.1e: TomVezo.com; 5.2a: Dr. Pusey, J. Williams, and J. Goodall, 1997, The influence of dom-
U. Nebelsiek/Peter Arnold, Inc.; 5.2b: Karl & Kay Ammann/ inance rank on the reproductive success of female chimpanzees,
Bruce Coleman Inc.; 5.2c: Alan Root/Survival Anglia/Oxford Science 277:828–831; 6.14: Photograph courtesy of Carola
Scientific Films; 5.3: Courtesy of Carola Borries (SUNY Stony- Borries; 6.15: From Figure 1 in J. B. Silk, S.C. Alberts, and J.
brook); 5.4: Norbert Wu/www.norbertwu.com; 5.6a: Robert Altmann, 2003, Social bonds of female baboons enhance infant
Boyd; 5.6b: Millard H. Sharp/Photo Researchers, Inc.; 5.6c: survival, Science 302:1331–1334; 6.16: Joan Silk; 6.17: Robert
Jacana/Photo Researchers, Inc.; 5.7a: K. G. Preston-Mafham/ Boyd and Joan Silk; 6.19: K. G. Preston-Mafham/Premaphotos
Premaphotos Wildlife; 5.7b: Courtesy of Susan Perry; 5.7c: Wildlife; 6.20a: © Art Wolfe/artwolfe.com; 6.20b: BIOS/Peter
Courtesy of Sue Boinski; 5.8a: Courtesy of Carola Borries; Arnold, Inc.; 6.21a: Robert Boyd; 6.21b: Robert Boyd; 6.22:
5.8b: Courtesy of Kathy West; 5.8c: Courtesy Marina Cords; Joan Silk; 6.23: From Figure 2 in P. H. Harvey and A. H. Har-
5.9: Courtesy of John Mitani; 5.10a: Joan Silk; 5.10b: Courtesy court, 1984, Sperm competition, testes size, and breeding sys-
of John Mitani; 5.11: Courtesy of John Mitani; 5.12: Karl and tems in primates, pp. 589–599, in Sperm Competition and the
Kay Ammann/Bruce Coleman Inc; 5.13: Joan Silk; 5.15: Joan Evolution of Animal Mating Systems, ed. R. L. Smith, Aca-
Silk; 5.16: Redrawn from Figure 5.13 in A. Richard, 1985, Pri- demic Press, New York; 6.24: From Figure 2 in P. H. Harvey
mates in Nature, W. H. Freeman, New York; 5.17a-b: From and A. H. Harcourt, 1984, Sperm competition, testes size, and
Figures 6.21 and 6.23 in R. D. Martin, 1990, Primate Origins breeding systems in primates, pp. 589–599 in Sperm Competi-
and Evolution, Princeton University Press, Princeton, N.J.; tion and the Evolution of Animal Mating Systems, ed. R. L.
5.18: From Figure 8.9 in J. G. Fleagle, 1988, Primate Adapta- Smith, Academic Press, New York; 6.25a: From Figure 2 in R.
tion and Evolution, Academic Press, San Diego, Calif. © Else- Palombit, in press, Infanticide and the evolution of pair bonds
vier Ltd. Reprinted by permission.; 5.19a: Courtesy of Lynne in nonhuman primates, Evolutionary Anthropology; 6.25b:
Isbell; 5.19b: Susan Perry; 5.19c: John Mitani; 5.19d: Joan Silk; Martin Harvey/Peter Arnold Inc.; 6.26: Erwin and Peggy Bauer/
5.19e: Carlão Limeira; 5.19f: Courtesy of Carola Borries; 5.- Wildstock; 6.27: Data from Figure 2 in P. A. Garber, 1997, One
20a-c: Adapted from Figure 4.1 in J. Terborgh, 1983, Five New for all and breeding for one: cooperation and competition as a
World Primates, Princeton University Press, Princeton, N.J.; tamarin reproductive strategy, Evolutionary Anthropology
5.21a: Courtesy of Susan Perry; 5.21b: courtesy of Lynne 5:187–199; 6.28: Erwin and Peggy Bauer/Wildstock; 6.29:
Isbell; 5.21c: Oxford Scientific/photolibrary; 5.23: Dieter and Joan Silk; 6.30: From Figure 2a in S. C. Alberts, J. C. Buchnan,
Mary Plage/Bruce Coleman Collection; 5.24: Courtesy of Car- and J. Altmann, 2006, Sexual selection in wild baboons: from
ola Borries; 5.25a: Robert Boyd; 5.25b: Robert Boyd; 5.25c: mating opportunities to paternity success, Anim. Behav. 72:
Wolfgang Kaehler; 5.25d: Robert Boyd; 5.25e: Courtesy of 1177–1196 © The Association for the Study of Animal Behav-
Lysa Leland; 5.26a-c: Robert Boyd; 5.27: From Figure 1, iour. Published by Elsevier Ltd. Reprinted by permission.; 6.31:
Schultz et al., 2004, A community level evaluation of prey From Figure 2b in S. C. Alberts, J. C. Buchnan, and J. Altmann,
behavioral and ecological characteristics on predator diet com- 2006, Sexual selection in wild baboons: from mating opportu-
position, Proc. R. Soc. Lond. B. 271: 725–732; 5.28: Joan Silk; nities to paternity success, Anim. Behav. 72: 1177–1196 © The
5.30: Courtesy of Sue Boinski; 5.31: Courtesy of Colin Association for the Study of Animal Behaviour. Published by
Chapman. Elsevier Ltd. Reprinted by permission.; 6.32: From Figure 2 of
C. Boesch, G. Kohou, H. Néné, and L. Vigilant, 2006, Male
CHAPTER 6 competition and paternity in wild chimpanzees of the Taï For-
est, Amer. J. Phys. Anthropol. 130: 103–115; 6.33: Photograph
6.0: Nick Gordon/naturepl.com; 6.1: Joan Silk; 6.2: henry-
by Ryne A. Palombit; 6.34: Joan B. Silk.
holdsworth.com; 6.3: Photograph by Kathy West; 6.4: Rudie
H. Kuiter, OSF/Animals Animals; 6.5: Photograph courtesy of
CHAPTER 7
Kathy West; 6.6: Ardella Reed Stock Photography; 6.7:
Adapted from data in A. Mori, 1979, Analysis of population 7.0: Courtesy of Susan Alberts; 7.1: K. G. Preston-Mafham/
changes by measurement of body weight in the Koshima troop Premaphotos Wildlife; 7.2: AP Photos; 7.3: Joan B. Silk; 7.8a:
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A19
Credits A19
Photograph courtesy of Kathy West; 7.8b: Joan Silk; 7.9a: Joan Part III: Bipedal footprints dating from
Silk; 7.9b: Joan Silk; 7.11a; Courtesy of Susan Perry; 7.11b:
approx. 3.5 mya discovered in Laetoli,
Courtesy of Marina Cords; 7.11c: Joan Silk; 7.11d: John
Mitani; 7.12: Adapted from Figure 4 in E. Kapsalis and C. M. Tanzania. (John Reader, 1999/Science
Berman, 1996, Models of affiliative relationships among free- Photo Library/Photo Researchers, Inc)
ranging rhesus monkeys (Macaca mulatta) I: criteria for kin-
ship, Behaviour 133:1209–1234; 7.13: Adapted from Figure 2
in F. Aureli, C. P. van Schaik, and J. A. R. A. M. van Hooff, CHAPTER 9
1989, Functional aspects of reconciliation among captive long-
9.0: Ronald C. Blakey/Colorado Plateau Geosystems, Inc.;
tailed macaques (Macaca fasicularis), American Journal of Pri-
9.1a-b: Figure from pp. 38–39 in R. J. G. Savage 1986, Mam-
matology 19:39–51; 7.14: Joan Silk; 7.15: From Figure 1 in
mal Evolution: An Illustrated Guide, © Facts on File, Inc., an
J. B. Silk, S. C. Alberts, and J. Altmann, 2004, Patterns of coali-
imprint of Infobase Publishing. Reprinted with permission of
tion formation by adult female baboons in Amboseli, Kenya,
the publisher; 9.2: From Figure 5.3 in R. D. Martin, 1990, Pri-
Animal Behaviour 67: 573–582; 7.16a: From Table 1 in J. A.
mate Origins and Evolution: A Phylogenetic Reconstruction.
Johnson, 1987, Dominance rank in juvenile olive baboons,
Reprinted by permission of Princeton University Press, Prince-
Papio anubis: the influence of gender, size, maternal rank and
ton, N. J.; 9.3: © Art Wolfe/artwolfe.com; 9.7: From p. 200 in
orphaning, Animal Behaviour 35:1694–1708; 7.16b: Joan Silk;
S. J. Jones, R. D. Martin, and D. A. Pilbeam, 1992, The Cam-
7.17: © Terry Whittaker; Frank Lane Picture Agency/Corbis;
bridge Encyclopedia of Human Evolution, © Cambridge Uni-
7.18: Joan Silk; 7.19: From Figure 1 in R. M. Seyfarth and D. L.
versity Press, Cambridge. Reprinted with the permission of
Cheney, 1984, Grooming, alliances, and reciprocal altruism in
Cambridge University Press; 9.8: Artwork by Doug M. Boyer
vervet monkeys, Nature 308:541–543.; 7.20: From Figure 2 in
from E. J. Sargis, 2002, Paleontology: Primate Origins Nailed,
F. B. M. de Waal, 1997, The chimpanzee’s service economy:
Science 298:1564, reprinted with permission; 9.10a-b: From
Food for grooming, Evolution & Human Behavior 18:375–
Figure 3 in K. D. Rose, 1994/5, The earliest primates, Evolu-
386; 7.21a: Marc D. Hauser; 7.21b: From Figure 2 in M. D.
tionary Anthropology 3:159–173, © Wiley-Liss, Inc., A Wiley
Hauser, M. K. Chen, and E. Chuang, 2003, Give unto others:
Company. Reprinted with permission of Wiley-Liss, Inc., a sub-
genetically unrelated cotton-tamarin monkeys preferentially
sidiary of John Wiley & Sons, Inc.; 9.11a-b: From Figure 11.9
give food to those who altruistically give food back, Proc. R.
in J. G. Fleagle, 1988, Primate Adaptation and Evolution,
Soc. Lond. B. 270:2363–2370.
Academic Press, © Elsevier Ltd. Reprinted by permission; 9.12:
From Figure 12.5 in J. G. Fleagle, 1988, Primate Adaptation
and Evolution, Academic Press, © Elsevier Ltd. Reprinted by
permission; 9.13: From Figure 12.6 in J. G. Fleagle, 1988, Pri-
CHAPTER 8
mate Adaptation and Evolution, Academic Press, © Elsevier
8.0: Courtesy of Susan Perry; 8.1: Cover: Time Life/Getty Ltd. Reprinted by permission; 9.14a-c: From Figure 1.10 in G.
Images: Photo: AP/Wide World Photos; 8.2: W. Perry Conway/ Conroy, 1990, Primate Evolution, W. W. Norton, New York;
Corbis; 8.3: Joan Silk; 8.4: Gallo Images/Corbis; 8.5: © Frans 9.15: From Figure 12.13 in J. G. Fleagle, 1988, Primate Adap-
Lanting/Corbis; 8.6: National Geographic/Getty Images; 8.7a: tation and Evolution, Academic Press, © Elsevier Ltd.
Joan Silk; 8.7b: Photograph courtesy of Susan Perry; 8.8: Wolf- Reprinted by permission; 9.17: Robert Boyd; 9.19a-c: From
gang Kohler, 1927, The Mentality of Apes. Routledge & Kegan Figure 2 in R. D. Martin, 1993, Primate origins: plugging
Paul, Ltd., London: Reproduced with permission from the the gaps, Nature 363:223–234; 9.20: Christophe Ratier/
publisher; 8.9a-c: Redrawn from p. 109 in S. J. Jones, R. D. NHPA.UK; 9.21: Photograph courtesy of Laura MacLatchy;
Martin, and D. A. Pilbeam, 1992, The Cambridge Encyclope- 9.22: From Figure 13.7 in J. G. Fleagle, 1988, Primate Adap-
dia of Human Evolution, Cambridge University Press, Cam- tation and Evolution, Academic Press, © Elsevier Ltd.
bridge; 8.10a: © Steve Bloom; 8.10b: Courtesy of Susan Perry; Reprinted by permission; 9.23: Figure 8.21 from R. Lewin and
8.11: Robert Boyd and Joan Silk; 8.12: Joan Silk; 8.14a: Joan R. Foley, 2004, Principles of Human Evolution, 2e, © Black-
Silk; 8.14b: Photograph courtesy of Susan Perry; 8.15: Courtesy well Science Ltd., a Blackwell Publishing company. Repro-
of E. Menzel; 8.16: Photo courtesy of Cognitive Evolution duced with permission of Blackwell Publishing Ltd.; 9.24:
Group, University of Louisiana at Lafayette; 8.18: From Figure From Figure 2.23 in R. D. Martin, 1990, Primate Origins and
1, J. I. Flombaum, L. R. Santos, 2005, Rhesus monkeys attrib- Evolution: A Phylogenetic Reconstruction. Reproduced by per-
ute perceptions to others. Curr. Biol. 15: 447–452, © 2005 mission of Princeton University Press, Princeton, N. J. (Rdg. 1–
Elsevier Ltd., with permission from Elsevier; 8.19: Figure 1 2); redrawn from pp. 50 and 51 in R. Potts, 1996, Humanity’s
from E. Hermann, J. Call, M. V. Hernandez-Lloreda, B. Hare, Descent, William Morrow, New York; 9.25: Photo courtesy,
and M. Tomasello, 2007, Humans have evolved specialized Salvador Moya-Sola; reproduced with permission of Nature
skills of social cognition: the cultural intelligence hypothesis, 379:156–159; 9.26: From S. Moya-Sola, M. Koehler, D. M.
Science 317:1360–1366; reprinted with permission from Alba, I. Casanovas-Vilar, and J. Galindo, 2004, Pieropithecus
AAAS. catalaunicus, a new middle Miocene great ape from Spain,
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A20
A20 Credits
Science 306:1339–1344. Figure 2; reproduced with permission Goodall, 1994, Patterns of predation by chimpanzees on red
from AAAS. colobus monkeys at Gombe National Park, 1982–1991, Amer-
ican Journal of Physical Anthropology 94:213–228; 10.42:
CHAPTER 10 Robert Boyd (Rdg. 1–3), courtesy of Getty Conservation Insti-
tute, Los Angeles, California, © 1999, the J. Paul Getty Trust;
10.0: Courtesy of M.P.F.T.; 10.1: From Figures 20.15 and
10.43: Courtesy of William McGrew; 10.44: Adapted from
20.20 in L. Aeillo and C. Dean, 1990, An Introduction to Evo-
Table 13 in C. Boesch and H. Boesch, 1989, Hunting behavior
lutionary Human Anatomy, Academic Press, New York; 10.3:
of wild chimpanzees in the Taï National Park, American Jour-
From D. C. Johanson and M. A. Edey, 1981, LUCY: The
nal of Physical Anthropology 78: 547–573; 10.45: Courtesy of
Beginnings of Humankind. © Luba Dmytryk Gudz/Brill
Craig Stanford.
Atlanta. Reprinted with permission; 10.4: Copyright M. P. F. T.;
photograph courtesy of Michael Brunet; 10.5: B. Senut et
al./C.R. Acad. Sci. Paris, 2001, Sciences de la Terre et des plan- CHAPTER 11
etes/Earth and Planetary Sciences 332: 141; 10.6a: From Fig- 11.0: Peter Johnson/Corbis; 11.1: From p. 113 in K. Schick and
ure 1 D: in Y. Haile-Selassie, G. Suwa, and T.G. White, 2004, N. Tooth, 1993, Making Silent Stones Speak, Simon and Schus-
Late Miocene teeth from Middle Awash, Ethiopia, and early ter, New York; 11.2: Redrawn from p. 141 in K. Schick and N.
hominid dental evolution, Science 303(5663): 1503–1505; Tooth, 1993, Making Silent Stones Speak, Simon and Schuster,
reprinted with permission form AAAS; 10.6b: From Figure 1 B New York; 11.4: Courtesy of Kim Hill; 11.5: Courtesy of
in Y. Haile-Selassie, G. Suwa, and T. G. White, 2004, Late Nicholas Blurton Jones; 11.9: Francesco d’Errico; 11.10: From
Miocene teeth from Middle Awash, Ethiopia, and early Figure 6.1 in R. Potts, 1988, Early Hominid Activities in Oldu-
hominid dental evolution. Science 303(5663):1503–1505. re- vai, © Aldine Publishers. Reprinted by permission of Aldine
printed with permission from AAAS; 10.7: Robert Boyd; 10.9: Transaction, a division of Transaction Publishers; 11.11a-c:
National Museums of Kenya; 10.10: Redrawn from Figure 1 in D. C. Johanson and M. A. Edey, 1981, LUCY: The Beginnings
M. G. Leakey, C. S. Feibel, I. McDougall, and A. Walker, of Humankind. © Luba Dmytryk Gudz/Brill Atlanta. Reprinted
1995, New four-million-year-old hominid species from with permission; 11.12: Adapted from Figure 3 in R. Potts,
Kanapoi and Allia Bay, Kenya, Nature 376:565–571; 10.11: 1984, Home bases and early hominids, American Scientist
Institute of Human Origins/Nanci Kahn; 10.12: John 72:338–347; 11.13: © Art Wolfe/artwolfe.com; 11.14a:
Reader/Science Photo Library/Photo Researchers, Inc; 10.13: A. National Museum of Natural History, Smithsonian Institution,
afarensis, “Lucy” reconstructed skeleton. Dr. Owen Lovejoy Washington, D.C. (Rdg. 1–2): courtesy of Kathy D. Schick and
and students, Kent State University. © 1985 David L. Brill. Nicolas Toth, Craft Research Center, Indiana University;
Reprinted with permission; 10.14: AFP/Getty Images; 10.15: 11.14b: National Museum of Natural History, Smithsonian
Figure 4.24 from R. G. Klein, 1989, 1999, The Human Career: Institution, Washington, D.C. (Rdg. 1–2): courtesy of Kathy D.
Human Biological and Cultural Origins, 2nd ed. © The Uni- Schick and Nicolas Toth, Craft Research Center, Indiana Uni-
versity of Chicago. Reprinted by permission of the University versity; 11.14c: National Museum of Natural History, Smith-
of Chicago Press; 10.17: From Figure 3a in Z. Alemseged et al., sonian Institution, Washington, D.C. (Rdg. 1–2): courtesy of
2006, A juvenile early hominin skeleton from Dikika, Ethiopia. Kathy D. Schick and Nicolas Toth, Craft Research Center, Indi-
Nature 443:296–301; 10.18: 1999, John Reader/Photo ana University; 11.15a: Y. Arthus-Bertrand/Peter Arnold, Inc.;
Researchers, Inc.; 10.19: From Figure 5 in Z. Alemseged et al., 11.15b: Y. Arthus-Bertrand/Peter Arnold, Inc.; 11.16: Robert
2006, A juvenile early hominin skeleton from Dikika, Ethiopia. Boyd and Joan Silk; 11.17: Joan Silk; 11.18: Robert Boyd;
Nature 443:296–301; 10.20: Figure 4.2 © 1999 by Kathryn 11.19: Robert Boyd; 11.20: Drawing by Christian Gurney
Cruz-Uribe from R. G. Klein, 1999, The Human Career: from Figure 2 in Henry T. Bunn and Ellen M. Kroll, 1986, Sys-
Human Biological and Cultural Origins, 2nd ed. © The Uni- tematic butchery by Plio/Pleistocene hominids at Olduvai
versity of Chicago. Reprinted by permission of the University Gorge, Tanzania, Current Anthropology 27:432–452; 11.21:
of Chicago Press; 10.22: A. garhi, BOU-VP-12/130, Bouri, Courtesy of Robert Bailey; 11.22: Figure 4.41 from R. G.
cranial parts, cranium reconstruction. Original housed in Klein, 1989, 1999, The Human Career: Human Biological
National Museum of Ethiopia, Addis Ababa. © 1999 David L. and Cultural Origins, 2nd ed. © The University of Chicago.
Brill. Reprinted with permission; 10.23: National Museums of Reprinted by permission of the University of Chicago Press;
Kenya; 10.24a-c: From p. 80 in R. Lewin, 1989, Human Evo- 11.23: Adapted from Figure 4 in R. Potts, 1984, Home bases
lution, Blackwell Scientific, Boston; 10.26: University of Wit- and early hominids, American Scientist 72:338–347.
watersrand; 10.27: Wolfgang Kaehler; 10.28: Maeve Leakey;
10.29: Alan Walker; 10.30: National Museums of Kenya;
CHAPTER 12
10.35: Robert Boyd; 10.36a-b: Joan Silk; 10.37a-b: Joan Silk;
10.38: Joan Silk; 10.39: From Figure 3.1 in R. I. M. Dunbar, 12.0: Photo courtesy of Prof. Peter Brown, The University of
1988, Primate Social Systems, Cornell University Press, Ithaca, New England, Armidale, Australia; permission also granted by
N.Y.; 10.40a-b: Courtesy of Craig Stanford; 10.41: Adapted Nature Publishing Group (http://www.nature.com); 12.2: Fig-
from Figure 1 in C. Stanford, J. Wallis, H. Matama, and J. ure 5.10, © Kathryn Cruz-Uribe from R. G. Klein, 1989, 1999,
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A21
Credits A21
The Human Career: Human Biological and Cultural Origins, Russian Plain, pp. 714–742 in The Human Revolution, ed. by
2nd ed. © The University of Chicago. Reprinted by permission P. Mellars and C. Stringer, © Edinburgh University Press.
of the University of Chicago Press; 12.3: Figures 4.2, © Kathryn Reprinted by permission of Princeton University Press, Prince-
Cruz-Uribe from R. G. Klein, 1989, 1999, The Human Career: ton, N.J.; 13.8: Figure 7.22, from R. G. Klein, 1989, 1999, The
Human Biological and Cultural Origins, 2nd ed. © The Uni- Human Career: Human Biological and Cultural Origins, 2nd
versity of Chicago. Reprinted by permission of the University ed. © The University of Chicago. Reprinted by permission of
of Chicago Press; 12.4: Alan Walker; 12.5: Photograph cour- the University of Chicago Press; 13.9: From Figure 13.4 in P.
tesy of Alan Walker, reprinted with permission of the National Mellars, 1996, The Neanderthal Legacy, Princeton University
Museums of Kenya; 12.6: From p. 232 in K. Schick and N. Press. Reprinted by permission of Princeton University Press;
Tooth, 1993, Making Silent Stones Speak, Simon and Schuster, 13.10: © Jean Clottes/DRAC Rhône-Alpes; 13.11: Ulmer
New York; 12.8: National Museums of Kenya; 12.12: David Museum, Ulm, Germany; 13.12: Redrawn from L. Excoffier,
Lordkipanidze, Georgian State Museum, Georgian Academy of 2002, Human demographic history: refining the recent African
Sciences; 12.13: National Anthropological Archive; 12.14: origins model, Current opinion in Genetics and Development
From p. 224 in S. Jones, R. Martin, and D. Pilbeam, 1992, The 12:675–682, Figure 1d; 13.16: From Figure 2 in F. Prugnolle, A.
Cambridge Encyclopedia of Human Evolution. Reprinted with Manica, and F. Balloux, Geography predicts neutral genetic
the permission of Cambridge University Press, Cambridge; diversity of human populations, Current Biology 15:159–160;
12.15: R. Potts, Smithsonian Institution; 12.16: Redrawn from 13.19: Redrawn from P. Forster, 2004, Ice ages and the chronol-
Figure 17 in C. Stringer and C. Gamble, 1993, In Search of the ogy of human dispersals: a review, Phil. Trans. R. Soc. Lond. B
Neanderthals, Thames & Hudson, New York; 12.18: Figures 359:255–264, Figure 2; 13.21: Housed in National Museum of
5.22 and 5.26, © Kathryn Cruz-Uribe, from R. G. Klein, 1989, Ethiopia, Addis Ababa, photo © 2001, David L. Brill/Brill,
1999, The Human Career: Human Biological and Cultural Ori- Atlanta; 13.22: Redrawn from Figure 15 in P. Mellars, 1989,
gins, 2nd ed. © The University of Chicago. Reprinted by per- Major issues in the emergence of modern humans, Current
mission of the University of Chicago Press; 12.19: Figure 6.24 Anthropology 30:349–385; 13.23: S. McBrearty and A. Brooks;
from R. G. Klein, 1989, 1999, The Human Career: Human 13.24: © Chip Clark Museum of Natural History, Smithsonian;
Biological and Cultural Origins, 2nd ed. © The University of 13.25: Figure 3 in A. Bouzouggar, et al., 2007, 82,000-year-old
Chicago. Reprinted by permission of the University of Chicago shell beads from North Africa and implications for the origins
Press; 12.21: Figure 5.32, © Kathryn Cruz-Uribe, from R. G. of modern human behvior, PNAS 104:9964–9969; 13.26: Cen-
Klein, 1989, 1999, The Human Career: Human Biological tre for Development Studies, University of Bergen.
and Cultural Origins, 2nd ed. © The University of Chicago.
Reprinted by permission of the University of Chicago Press;
12.24: Javier Trueba/Madrid Scientific Films; 12.25: Adapted
Part IV: Footprints on the moon from the
from Figure 1, Greenland Ice-Core Project, 1993, Climate
instability during the last interglacial period recorded in the Apollo landing (NASA)
GRIP ice core, Nature 364:203–207; 12.26: Figure 6.48, ©
Kathryn Cruz-Uribe, from R. G. Klein, 1989, 1999, The CHAPTER 14
Human Career: Human Biological and Cultural Origins, 2nd
14.0: Ariel Skelley/Corbis; 14.1a-c: From Figure 20.12 in W.
ed. © The University of Chicago. Reprinted by permission of
Bodmer and L. L. Cavalli-Sforza, 1976, Genetics, Evolution,
the University of Chicago Press; 12.28a-b: From Figure 34 in C.
and Man, W. H. Freeman, New York; 14.2: AP/Wide World
Stringer and C. Gamble, 1993, In Search of the Neanderthals,
Photos; 14.3a: George Long/Sports Illustrated; 14.3b: Richard
Thames & Hudson, New York; 12.30a: BIOS/Peter Arnold,
Mackson/Sports Illustrated; 14.4: Adapted from p. 4 of M. Gop-
Inc; 12.30b: Maurizio Lanini/Corbis; 12.31: Martin Harvey;
nik and M. B. Crago, 1991, Familial aggregation of a develop-
12.32: Erik Trinkaus; 12.33: Joan Silk; 12.34: Eric Delson;
mental language disorder, Cognition 39:1–50; 14.5a: Courtesy
12.35: Photograph by Michael Day.
of Marion I. Barnhart, Wayne State University Medical School,
Detroit, Michigan; 14.5b: Courtesy of Marion I. Barnhart,
CHAPTER 13
Wayne State University Medical School, Detroit, Michigan; 14.-
13.0: © Jean Clottes/DRAC Rhône-Alpes; 13.1: Figure 7.2, 7a-b: Adapted from Figures 2.14.1B and 2.14.9 in L. L. Cavalli-
from R. G. Klein, 1989, 1999, The Human Career: Human Sforza, P. Menozzi, and A. Piazza, 1994, The History and
Biological and Cultural Origins, 2nd ed. © The University of Geography of Human Genes, Princeton University Press, Prince-
Chicago. Reprinted by permission of the University of Chicago ton, N.J.; 14.8: Sarah Errington/Panos Pictures; 14.9a-b: From
Press; 13.3: Courtesy of AAAS; 13.4: Jean-Michel Labat/AUS- Figure 1. in A. Beja-Pereira, G. Luikartz, P. R. England, D. G.
CAPE International; 13.5: From Figure 5.6 in F. Bordes, 1968, Bradley, O. C. Jann, G. Bertorelle, A. T. Chamberlain, T. P.
La Paléolithique dans le monde, Hachette, Paris; 13.6: Adapted Nunes, S. Metodiev, N. Ferrand, and G. Erhardt, 2003, Gene-
from Figure 7.9 in R. Klein, 1989, The Human Career, Univer- culture coevolution between cattle mik protein genes and human
sity of Chicago Press, Chicago; 13.7: From Figure 34.8 in O. lactase genese, Nature Genetics, 35(4):1–3; 14.11: Redrawn
Soffer, 1989, The Middle to Upper Paleolithic transition on the from Figure 100 in G. Flatz, 1987, The genetics of lactose
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A22
A22 Credits
digestion in humans, Advances in Human Genetics 16:60.; From Figures 929 and 943 in J. Needham, 1988, Science and
14.12: Jean Higgins/Unicorn Stock; 14.13: From F. Prugnolle, A. Civilization in China, vol. 4, pt. 3, Cambridge University Press,
Manica, and F. Balloux, 2005, Geography predicts neutral Cambridge, UK; 15.12: Courtesy of Susan Perry; 15.13: Ken-
genetic diversity of human populations, Current Biology, 5:159– nosuke Tsuda/Nature Production, Tokyo; 15.14: Behavioral
160, © Elsevier Ltd. All rights reserved. Reprinted by permis- Ecology Research Group, University of Oxford; 15.15: © Pete
sion; 14.14: From A. Manica, F. Prugnolle, and F. Balloux, 2005, Oxford/Minden Pictures; 15.16: Courtesy of William McGrew;
Geography is a better determinant of human genetic diversity 15.17: Anup Shah/Nature Picture Library; 15.18: Photograph
than ethnicity, Human Genetics 118:366–371, with kind per- by Gaku Ohashi, Primate Research Institute, Kyoto University;
mission from Springer Science+Business Media, © Springer 15.19: Robert Boyd; 15.20: Corbis; 15.21: Kennan Harvey.
Berlin/Heidelberg; 14.15: From Figure 1, p. 241, in D. Hatl,
1980, Principles of Population Genetics, Sinauer, Sunderland, CHAPTER 16
Mass.; 14.16: John W. McDonough; 14.17: From Figure 19.12
16.0: The Marriage Contract, Jan Jozef II Horemans, 1768.
in W. F. Bodmer and L. L. Cavalli-Sforza, 1976, Genetics, Evo-
Rockoxhouse, Antwerp; 16.1: English Heritage Photo Library/
lution, and Man, Freeman, New York; 14.19: From Figure 16.6
G. P. Darwin on behalf of Darwin Heirlooms Trust; 16.2:
in G. A. Harrison, J. M. Tanner, D. R. Pilbeam, and P. T. Baker,
From Tables 7.1 and 13.3 in N. Howell, 1979, Demography of
1988, Human Biology: An Introduction to Human Evolution,
the Dobe !Kung, Academic Press, New York; 16.8a-b: From
Variation, Growth, and Adaptability, 3rd ed., Oxford Science
Tables 2 and 6 in D. Buss, 1989, Sex differences in human mate
Publications, Oxford; 14.20: From Figure 2.13.4 in L. L.
preferences: evolutionary hypotheses tested in 37 cultures,
Cavalli-Sforza, P. Menozzi, and A. Piazza, 1994, The History
Behavioral and Brain Sciences 12:1–49; 16.9: Photograph cour-
and Geography of Human Genes, Princeton University Press,
tesy of Monique Borgerhoff Mulder; 16.10: From Figure 3.2 in
Princeton, N.J.; 14.21a-b: From Michael Bamshad, et al., 2004,
M. Borgerhoff Mulder, 1988, Kipsigis bridewealth payments,
Deconstructing the relationship between genetics and race,
pp. 65–82 in Human Reproductive Behavior, ed. by L. Betzig,
Nature Reviews Genetics 5:598–609. Reprinted by permission
M. Borgerhoff Mulder, and P. Turke, Cambridge University
from Macmillan Publishers Ltd. © Nature Publishing Group.
Press, Cambridge; 16.11: Photograph courtesy of Nancy
Levine; 16.12: From N. L. Levine and J. B. Silk, 1997, When
CHAPTER 15
polyandry fails: sources of instability in polyandrous mar-
15.0: © Perry van Duijnhoven; 15.1: Stephen Dalton/Photo riages, Current Anthropology 38:375–398; 16.15: Courtesy of
Researchers, Inc; 15.2: Joe Cavanaugh/DDB Stock Photo; 15.3: Frank Marlowe; 16.17: From Figure 6.10 in J. Wood, 1994,
Oil painting by Martin Pate; courtesy of Cultural Resources, Dynamics of Human Reproduction, Aldine de Gruyter, New
National Park Service; 15.4: Modified from Figure 6.6 in W. H. York; 16.18: Photograph by Lennart Nilsson/Albert Bonniers
Durham, 1992, Coevolution, Stanford University Press, Stan- Forlag AB, Behold Man, Little, Brown and Company; 16.19:
ford, Calif.; 15.5: Anup Shah/Nature Picture Library; 15.6a: Stephen Dalton/Photo Researchers, Inc; 16.20: Adapted from
Adapted from Table 12.6 in A. P. Wolf, 1995, Sexual Attraction Table III in J. B. Silk, 1980, Adoption in Oceana, American
and Childhood Association, Stanford University Press, Stan- Anthropologist 82:799–820.
ford, Calif.; 15.6b: Adapted from Tables 11.1, 11.2, and 11.3
in A. P. Wolf, 1995, Sexual Attraction and Childhood Associ- EPILOGUE
ation, Stanford University Press, Stanford, Calif.; 15.7: K. G.
1: By courtesy of the National Portrait Gallery, London.
Preston-Mafham/Premaphotos Wildlife; 15.8: Dr. Ronald H.
Cohn/The Gorilla Foundation; 15.9: Steve Winters/courtesy of
APPENDIX
the Language Research Center; 15.10: Adapted from Table 71
in J. Needham, 1988, Science and Civilization in China, vol. 4, 1: Adapted from Figure B.1.b–c in G. Conroy, 1990, Primate
pt. 3, Cambridge University Press, Cambridge, UK; 15.11a-b: Evolution, W. W. Norton, New York.
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A23
INDEX
Page numbers in italics refer to illustrations, tables and figures.
abductor muscles, 249, 250 anatomically modern humans in, grooming and, 186–88, 186, 193–
Abkazov, Arha, 62 342–43, 358–59, 363–64, 367, 95
Aché, 287–88, 287, 289 368, 370, 373–74 Hamilton’s rule and, 182–84, 183,
Acheulean tool industry, 310–12, anthropoid evolution in, 237, 237 192
310, 318, 323, 348–49 deforestation in, 144 natural selection and, 176–78,
achondroplasia, 12 emergence of H. sapiens in, 335– 193–94
activator protein, 47–48, 47, 48 39, 336, 339 in primates, 176–80, 179, 181,
Adapidae, 230–31, 231, 232 and evolution of modern human 191–92, 193–95, 194, 195
adaptation, 2–22, 71–82, 72, 73, 74, behavior, 383 reciprocal, 192–94, 193, 194, 195,
77, 80, 82 H. heidelbergensis in, 322–24, 322 195
behavior and, 74, 80, 82 hominid diversification in, 253–70, social behavior and, 177, 179,
competition and, 11–12 254 181, 183, 184, 185, 186, 187,
complex, 12–17, 79 hominoid evolution in, 240–44 188, 189, 190, 193, 194, 195
constraints on, 71–80 Kalahari Desert of, 438, 439 see also kin selection
in convergent evolution, 104–5 Miocene climate of, 246–47 Amboseli National Park, Kenya, 94,
correlated characters and, 72–74 patterns of genetic variation in, 157, 159, 167, 207, 275–76, 275
culture as, 437–41, 438, 439 403–4, 404 American Sign Language, 429–30,
before Darwin, 2–4, 11 Pleistocene climate of, 319–21 430
Darwin on, 4–9, 13–14, 19, 61, primates in, 122, 122, 143 amino acids, 42–44, 43, 103, 103,
63–64, 148, 220 African Americans, 42, 410 104, 346, 390, 391
developmental constraints in, 80 Afrikaners, 402 codon variants of, 44
dietary, 133, 133 aggression, redirected, 207–8 in primate nutrition, 128
disequilibrium and, 74–75, 74 aging, see senescence proteins in sequence of, 42–44, 43
eyes and, 3, 15–16, 15 agriculture, 382–83, 402, 420 tRNA attached to, 44
genetic drift and, 79 alanine, 44 aminoacyl-tRNA synthetase, 45
genetics in, 59–60, 61–63, 73–74, alarm calling, 178–80, 179, 181 Amish, 401–2, 401
75–79, 77, 78, 80–82 Alberts, Susan, 169 analogous characters, 105
individual selection and, 11–12 Alemseged, Zeresenay, 256, 260 analogy, reasoning by, 117
language as, 382, 429, 430 Algeripithecus minutus, 233 ancestral characters, 103, 104–7,
local vs. optimal, 79–80 alleles, 29, 59 105–6
optimal, 79–80 dominant, 30 Anderson, Kermyt, 459–60
postulates of, 5–6, 7–8, 11 gene frequency and, 54–56 angiosperms, 227
by primates, 117–18 in linkage, 32–36 Antarctica, 122, 222, 223, 232
speciation in, 5–6, 10, 12–14 ratio of, 30–35, 33, 35 anthropoid primates (Anthropoidea),
variation and, 5–6, 12–17 in recombination, 32–36 121
in variation maintenance, 64–68 allopatric speciation, 91–94, 92, 93, evolution of, 231–37, 233, 234
adaptive radiation, 96, 96 95, 339 life history of, 202
adenine, 38, 39 Altamira, Spain, 353 in taxonomy of living primates,
adoption, 466–70 Altmann, Jeanne, 185 122–26, 123
as form of altruism, 466–70, 468, altruism: anticodons, 44
469 adoption as form of, 466–70, 468, apatite crystals, 226
in industrialized societies, 468–70 469 apes, 119, 125, 130, 154, 240, 240
kin selection and, 466–68 alarm calling and, 178–80, 179, early evolution of, 233–37, 235,
in Oceania, 466–70, 468, 469 181 242–44, 242
Aegyptopithecus zeuxis, 233, 235, in behavior, 177, 179, 181, 183, foraging by, 287, 287
241 184, 185, 186, 187, 188, 189, intelligence of, 207, 209, 213, 214
Afar depression, Ethiopia, 256 190, 193, 194, 195, 290 observational learning by, 437–38
affiliative contacts, 186 coalition formation and, 177–80, Apidium phiomense, 233
Africa, 86, 177–78, 221, 222, 232, 179, 188–91, 189, 192–95, 209, arboreal primates, 100–101
233, 253, 367, 436 209 Arcy-sur-Cure, France, 381, 381, 382
A23
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:23 PM Page A24
A24 Index
Ardipithecus kadabba, 248, 252, 252 postcranial skeleton of, 262 culture and, 431–44, 432, 434,
Ardipithecus ramidus, 252–53, 253, skull of, 262, 262 438, 441
268, 271 teeth of, 262, 262 environmental interactions and,
area-volume ratios, 81, 81 Awash Basin, Ethiopia, 252, 256, 418–19
argon-argon dating, 226, 268, 317, 264, 282 evolution and, 68–71, 69, 70,
373 aye-ayes, 118, 119 417–41, 420, 423, 424, 432,
art: 434, 438, 441
of anatomically modern humans, baboons, 145, 148 and evolution of modern humans,
349, 357–58, 357, 358, 379–80, chacma, 168 380–83
379 coalitions among, 177–78, 189 as human revolution, 376, 382–
cave, 357–58, 358, 379 ecology of, 127, 138, 140 83
of Middle Stone Age, 379, 379 food gathering by, 276, 276 in inbreeding avoidance, 422–26,
of Upper Paleolithic, 357–58, 358 gelada, 119, 167, 167, 203 423, 424, 425
artificial selection, 19, 20, 22, 74 hamadryas, 161 maternal, 160–61, 161
Asfaw, Berhane, 252, 264 infanticide by, 172–73 in Middle Stone Age, 376–82
Asia, 223, 330 intelligence of, 207, 208–9 morphology and, 68
anatomically modern humans in, mating systems of, 154, 155, 155, natural selection and, 68–71, 69,
350 158–60, 161, 161, 163, 167, 70, 95, 419–20
deforestation in, 144 167, 168, 168, 169, 171, 172– nature-nurture question and, 418–
primates in, 122, 122, 143 73, 172, 173 19
Upper Paleolithic tool tradition in, observational learning by, 438 observational learning and, 433–
350 predation of, 138, 139 34, 434
Aterian tool industry, 378, 378 savannah, 129, 171, 172, 439 of primates, 117, 118–21, 125,
Atlantic dog whelk, 15 sexual dimorphism of, 163 135–38, 179, 181, 209–14, 211
Atsalis, Sylvia, 154 social behavior of, 185, 186, 187, reasoning ability and, 421
Aureli, Felipo, 187–88 189, 189, 190, 193 reconciliatory, 187–88, 187
Aurignacian tools, 351–52, 352, 375, social habits of, 158–60, 160 sexual selection and, 161, 165–67
375 social organization of, 280 social, see social behavior
aurochs, 19, 330, 333 bachelor males, 164 social facilitation and, 433, 434
Austad, Steven, 202 Bacho Kiro, Bulgaria, 352 speciation and, 86–87
Australia, 122, 222, 232, 330 Backwell, Lucinda, 293 symbolic, 349
anatomically modern humans in, Bailey, Shara, 381 behavioral flexibility hypothesis, 203,
342, 349–50 balanced polymorphism, 393 204–5
Lake Mungo, 349 calculating gene frequencies for, Beja, 397, 397
marsupial fauna of, 16, 17 394, 394 Beja-Pereira, Albano, 398
Australopithecus (australopithecines), bald eagles, 151 Belanger’s tree shrew, 220
254, 263 bamboo, 319 “benefit,” use of term, 150
A. anamensis, 255–56, 255, 268, Bamshad, Michael, 414 Berelekh, 350
271, 273, 309 Barton, Robert, 155 Bergman, Thore, 208
A. bahrelghazali, 271, 271 basal metabolic rate, 127–28, 127 Berman, Carol, 186
A. garhi, 253, 264–65, 264, 285 bases, 38 Beyrich’s split shell, 15
A. rudolfensis, 268, 271, 285, 286 basicrania, 332 biface tools, 318–19
maturation rate of, 263–64 Batadomba-lena Cave, 350 bilateral symmetry, 131
teeth of, 263, 263 Bateson, William, 61 Binford, Lewis, 333
Australopithecus afarensis, 253, 255, bats, 96, 97, 97, 219, 227, 233, binocular vision, 121
256–61, 263–64, 266, 271, 271, 326 biochemical pathways, 41
273, 274–75, 276 Beagle, HMS, 5, 5, 444 biological species concept, 87–88, 95
arboreality of, 260 Beehner, Jacinta, 173, 208 biology, culture and, 441
bipedalism of, 259, 274–75 behavior, 50, 71, 95 bipedalism, 220, 221, 243
modern humans compared with, adaptation and, 74, 80, 82 in A. afarensis, 249, 259, 274–75
249 altruism in, 177, 179, 181, 183, in A. africanus, 262
sexual dimorphism in, 260–61 184, 185, 186, 187, 188, 189, in chimpanzees, 274–75, 275
skull of, 256–57 190, 193, 194, 195, 290 evolution of, 273–75, 274, 275
teeth of, 257, 258, 263, 276 of anatomically modern humans, heat stress and, 273–75, 274
Australopithecus africanus, 253, 343, 377–83 in hominids, 247, 249, 250, 252–
261–64, 262 in archaeological record, 348–58 53, 255, 259, 262, 266, 273–75,
bipedalism of, 261–62 artistic, 349 274, 275
brain of, 261–62 of chimpanzees, 346 in P. robustus, 266
maturation rate of, 263–64 complexity in, 376 torque in, 250
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A25
Index A25
birds, 111, 136, 150, 151, 152, 165, Broken Hill (Kabwa) cranium, 322, canalized vs. plastic, 69–70
227, 353 323, 325 complex, 20–21
bison, 333, 333, 353 Brooks, Alison, 376, 377, 380 correlative, 72–74, 74
“Black Skull,” 265, 265 Broom, Robert, 253, 266 in defining primates, 118–21, 119,
blades, 351, 351, 352, 377 Brues, Alice, 414–15 120, 121, 220, 230–31
blending inheritance, 21–22, 64–65 buffalo, 96, 377 derived, 103, 105–7, 105–6, 118–
Blind Watchmaker, The (Dawkins), Bulgaria, 352 21, 431–34, 432, 434
13 Burger, Joachim, 398 disequilibrium and, 74–75
Bloch, Jonathan, 228 bushbucks, 277, 277 evolution of, 70–71
Blombos Cave, South Africa, 378, bushpigs, 277, 377 of finches, 10, 10, 12, 18–19, 18,
379, 379 Buss, David, 446–47, 450–51 21–22
Bloom, Paul, 428 Byrne, Richard, 206, 211 maladaptive, 73
blue monkeys, 11, 145, 168, 186, in Mendelian genetics, 28
186 Caledonian crows, 435, 435 of minnows, 19–20
Blumenbach, Johann, 117 California abalone, 15 observable, 29
Blurton Jones, Nicholas, 288 Call, Josep, 211–12 outgroups and, 107
Boesch, Christophe, 169, 277 Callicebus monkeys, 185 phenotype and expression of, 63–
Boesch-Achermann, Hedwige, 277 Callitrichidae, 166 64
Boise, Charles, 267 camera-type eyes, 79–80, 80 in rates of evolutionary change,
bola, 285 Canada, 330 17–21, 18
Bolivia, 235 canalized environment, 419 recombination of, 32–36
bonnet macaque, 153, 184, 209 canines, 131, 234, 234, 251, 252, sexual selection and, 161
bonobos (Pan paniscus), 98, 110, 125 257, 258, 262, 262, 267 shared by primates, 118–21, 119,
language-trained, 430 Cape Town, South Africa, 460–61 120, 121
phylogeny of, 110, 111 capuchin monkeys, 124, 209, 209, in speciation, 98–99
sexual dimorphism in, 261 241, 433 Châtelperronian tool industry, 375
social behavior of, 126, 205 ecology of, 134, 136 checkerspot butterflies (Euphydryas
see also chimpanzees, common social behavior of, 186, 186 editha), 90, 90, 91
Borgerhoff Mulder, Monique, 454–56 tool use by, 434 cheetahs, 299, 299
Borneo, 202 white-faced, 169, 171 chemistry, 80
Borries, Carola, 157 carbohydrates, 129, 130 Cheney, Dorothy, 194, 207, 208, 429
Bose Basin, China, 318, 319 carbon-14 (radiocarbon) dating, 226, chickens, domestic, 104, 104
Botswana, 139, 172, 172, 208, 288, 349 chimerism, 191
377, 448 Carpolestes simpsoni, 228–30, 229 chimpanzees:
Boussou, Guinea, 437, 437 Carroll, Scott, 69–71 brain of, 251
Bouzouggar, Abdeljalil, 379 Cartmill, Matt, 229 genomes of, 345–48
Boyer, Doug, 228 catalysts, 38 human lineages and, 345–47
brain, brain size, 21, 230, 420 see also enzymes intelligence of, 213
of A. africanus, 262 Cavalli-Sforza, L. L., 402–3 mating systems of, 169, 170, 171,
of anatomically modern humans, cave art, 349, 357–58, 358, 379 178
21, 205 cave bears, 353 predation by, 140
complex foraging and, 291–92 cebus monkeys, 435, 435 taxonomy of, 125
foraging and, 291–92 cells: teeth of, 251
heat stress and, 273, 274 chromosomes in, 27, 27 tool use by, 206
of H. ergaster, 305, 309, 323 division of, 27–28 chimpanzees, common (Pan
of hominids, 247, 262, 264, 273 mitochondria in, 360 troglodytes), 86, 86, 98, 99, 100,
intelligence and, 205 nucleus of, 27, 27 100, 110, 148, 243, 246, 251, 265,
language and, 382 Cenozoic era, 219, 219, 223 361
and life span, 198–99 Central America, 19, 122, 122, 144, arboreality of, 260
of mammals, 205 232 bipedalism in, 274–75, 275
of Neanderthals, 330 cercopithecines, see Old World classification of, 110–12
of Paranthropus, 265 monkeys dwarf, see bonobos
in primate intelligence, 120, 205 chacma baboons, 168 food sharing by, 279–80, 279,
of prosimians, 205 Chad, 248, 251, 254, 256–57 280, 290, 291
Brain, C. K., 314 Chapman, Colin, 144 foraging by, 287, 287
Brain, Robert, 293 character displacement, 93–94 genetic variation in, 361, 361
Bramble, Dennis, 310 characters (traits): genome of, 345–46
Brazil, 145, 412 analogous vs. homologous, 105 grooming posture of, 436, 436
brideswealth, 454–56, 454, 455 ancestral, 103, 104–7, 105–6 hunting by, 277, 277, 290, 291
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A26
A26 Index
inbreeding and, 422, 424, 424 codons, 42, 44, 45 definition of, 431
intelligence of, 210–12, 211, 212 coefficient of relatedness, 182–84, as derived trait, 431–34, 432, 434
knee joint of, 250, 250 182, 183 end of human evolution and, 470
language-trained, 430 collagen, 42 environmental variation and, 388
mating system of, 159, 160 collected foods, 286–87 evolution and, 431–41, 432, 434,
maturation rate of, 263 colobus monkeys, 145, 251 438, 439
morphology of, 86, 86 black-and-white, 164 incest prohibition in, 424
observational learning in, 437–38 ecology of, 138, 140 language and, 426–27
phylogeny of, 98, 99, 100, 101, mating system of, 164 mate selection and, 451–52, 460
108, 108, 110–11, 110, 111 red, 134, 140, 277, 277 memes and, 440–41
postcranial skeleton of, 249 combinational control of gene in nonhuman primates, 433, 434
predation by, 138, 139, 140, 277, expression, 48, 49 observational learning and, 433–
277 comparative method, 100, 102, 118 34, 434, 438
pygmy, see bonobos competition, sexual dimorphism and, parenting and, 458–59, 460, 461
reproduction strategies of, 126 163–65 primates and, 433
sexual dimorphism in, 263–64 complex adaptations, 12–17 race concept and, 414–15
skull of, 258 complex characters, 20–21 social facilitation and, 433, 434
social behavior of, 187, 193–94, compound eyes, 79–80, 80 and unpredicted evolutionary
193, 194 computed tomography (CT scan), outcomes, 439–40
social organization of, 204, 205, 381 Currat, Matthias, 372
277–78, 279–80 Congo, 377–78, 378 Curtis, Garness, 317
and survival after disabilities, 336, conspecifics, 118 cusps, 131, 131, 133, 133
336 contest competition, 143 Cuvier, Georges, 117
teeth of, 258 contextual cues, 185 cystic fibrosis (CF), 387, 387, 392
tool use by, 86, 277, 278, 279, continental drift, 221–23, 222 cytoplasm, 45
287, 433, 435–37, 436 continuous variation, 12–13, 22, cytosine, 38, 39
China, 350, 432, 452 60–64
Bose Basin, 319 Convention of Trade in Endangered Dagosto, Marian, 229
Yinkou, 325 Species of Wild Flora and Fauna Daly, Martin, 463, 465
Zhoukoudian, 318, 325 (CITES), 145 Daphne Major Island, 7–8, 7, 9, 11,
chorion (membrane), 191 convergence, 16–17 18, 62, 64, 72, 72, 74, 93
chromosomes, 27, 27, 29, 62, 75, convergent evolution, 104–5, 271 Dart, Raymond, 261–62, 262
345, 359, 360, 360, 391 adaptation in, 104–5 Darwin, Charles, 60, 161, 444–45,
in cell division, 27–28 Copper Eskimos, 438 445, 474, 474
definition of, 27 coprolites, 318 on adaptation, 4–9, 13–14, 19, 61,
in diploid organisms, 27–28 Cords, Marina, 186 63–64, 148, 220
genes on, 28–29, 33 cores, 283, 283, 284, 285, 324, 324 background, 4–5
homologous pairs of, 28 deep-sea, 224, 224 on classification of humans, 117
in Mendel’s experimental results, corn, 67 on evolution, 4–6, 11, 13, 15–16,
28–31 corpus, luteum, 464 17, 21–22, 60, 86
proteins in, 37 correlative characters, 72–74, 74 on natural selection, 5–6, 21–22
repeated DNA sequences on, 50 Correns, Carl, 26 postulates of, 5–6, 7–8, 11, 180
Y, 359, 364 Cosmides, Leda, 421, 426 on species, 11, 21, 86
Chumash, 438 “cost,” 150 variation explanation difficult for,
cladistics, 110–12, 111 Costa Rica, 134, 433 21–22
Clark, J. Desmond, 285 Crick, Francis, 37 Darwin, Emma Wedgwood, 445, 445
Clarke, Ronald, 257 Croatia, 371 Darwin’s finches, see finches
cleavers, 310 Crockett, Carolyn, 168 Dasser, Verena, 207
climate and climate change, 232, 306, crocodiles, 111, 138, 139 Dawkins, Richard, 13–14, 440
327, 329–30, 337, 339 Cro-Magnon rock shelter, 343 Dean, Christopher, 263, 310
in fossil record, 221–23, 223 crosses, in pea plants, 25 deep sea cores, 224–25, 329
in Miocene, 246–47 crowned hawk eagles, 138–40, 139, deer:
of Pleistocene epoch, 306, 319–21 140 fallow, 333
seasonality of, 275–76, 275, 276 crural index, 332, 332 red, 19, 333, 333, 353, 353
clothing, 354–55 cuckoos, 467, 467 deforestation, 144–45, 144
Clutton-Brock, Tim, 164 culture, 382 dental formula, 131, 131
coalition (alliance) formation, 177– as adaptation, 437–41, 438, 439 deoxyribonucleic acid, see DNA
80, 188–91, 189, 190–91, 193–95 behavior and, 431–44, 432, 434, derived characters, 103, 105–7, 105–
intelligence and, 209, 209 438, 441 6, 118–21, 431–34, 432
coding sequences, DNA in, 39, 347 biology and, 441 d’Errico, Francesco, 293
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A27
Index A27
Descent of Man, The (Darwin), 117 Drakensberg Mountains, South eukaryotes, 45–46
Dettwyler, Katherine, 335 Africa, 94 Euler, Harald, 462–63
de Vries, Hugo, 26–27, 61 Drosophila (fruit flies), 28, 49, 345 Eurasia:
de Waal, Frans, 187, 194 Dryopithecus, 242, 243 H. erectus in, 315, 320–21, 325–
diabetes, 393, 395 Dubois, Eugene, 317, 317 27
Diamond, Jared, 393 duck-billed platypus, 103, 104, 104 H. heidelbergensis in, 322–24,
Diana monkeys, 140, 141 dugongs, 97, 97, 98 322, 325, 328
diastema, 257 Dunbar, Robin, 167–68, 187, 204 Pleistocene climate of, 319–21
Didelphis virginiana, 200 spread of hominids to, 337–38
Dikika child skeleton, 256, 257, 257, eagles, crowned hawk, 138–40, 139 evolution:
259, 259, 260 ecological intelligence hypothesis, of anthropoid primates, 231–37,
dinosaurs, 96, 218–19 203–4, 204 233, 234, 242–44, 242
diploid organisms, 27–28, 29 ecological species concept, 86–91 behavior and, 68–71, 69, 70, 417–
discontinuous variation, 12, 22 Efe, 301 41, 420, 423, 424, 432, 434,
Discover, 414–15 Egypt, 424 438, 441
discriminative parental solicitude, Fayum depression of, 233, 233, of characters, 72–74
463, 465 235, 237 constraints on, 80–82
disequilibrium, 74–75, 74 Ehrlich, Paul, 90 convergent, 104, 271
dizygotic (fraternal) twins, 405–6 eland, 377 culture and, 431–41, 432, 434,
Dmanisi, Republic of Georgia, 315– electron-spin-resonance dating, 226, 438, 439
17, 316 326, 374 Darwin on, 4–6, 11, 13, 15–16,
DNA (deoxyribonucleic acid), 29, elephants, 96, 199, 201, 201, 219, 17, 21–22, 60, 86
37–50, 65, 343, 364–65, 390, 398, 294, 312, 313–14, 333, 415 end of, 470
413 dwarf, 326 human dietary needs and, 74–75,
alternative splicing of, 46, 47 morphological restraints on, 80– 75
enzymes and, 40, 41 82 idea of progress and, 221
in eukaryotes, 45–46 stone tools and butchering of, 293 of locomotion in apes, 100, 100,
in genes, 37–40, 345–46 elk, see red deer 101
mitochondrial, 369 Enattah, Nabil, 397 mate preference and, 447–52
natural selection and, 48, 398– endocranial volume, 257 molecular basis of, 37–38, 50
401, 400 Engh, Anne, 158 parallel, 271, 272
of Neanderthals, 371–73, 371 environment, 122 random mating and, 59
noncoding (sequences), 46, 347– behavioral interaction with, 418– rates of change in, 17–21
48 19 sexual reproduction and, 59
in origin and spread of canalized vs. plastic, 419 terms used in, 149–50
anatomically modern humans, in fossil record, 221–23 evolutionary systematics, 110–12,
360–73 human reasoning and, 421 111
in prokaryotes, 44–45 natural selection and, 60 Excoffier, Laurent, 372
protein codes of, 42–44, 43, 45, environmental covariation, 405 executive brain, 205
345–47 environmental variation, 63 exons, 46
replication of, 38, 40, 41 environment of evolutionary extractive foods, 203, 286–87
sequences of, 39, 47, 50 adaptedness (EEA), 421 extractive foraging hypothesis, 204
structure of, 37–40, 39 enzymes, 39, 40 extra-pair copulations, 165–66
transposable elements of, 345 DNA and, 40, 41 eyes:
Dobzhansky, Theodosius, 61, 417 in regulation of genes, 40, 41 adaptation and, 3, 15–16, 15
dogs, 19, 22 see also catalysts compound vs. camera-type, 79–80,
domestication of, 355 Eocene epoch, 230–33, 230, 235, 237 80
hidden variation in, 67–68, 67 Eosimias, 233 of C. simpsoni, 229–30, 229
wild, 299 equilibrium, among populations, 10 evolution of, 17, 20–21, 79–80, 80
wolves as ancestors of, 19, 22, 22, Erus, Justus, 268 of fish, 3, 17
67–68, 67 Escheria coli, 44, 46, 48, 49 of modern humans, 3, 17, 20–21,
dolphins, 121, 121 essentialism, 11 79–80, 80, 382
dominance hierarchies, 187, 208 estrus, 165 of primates, 120, 121, 228, 228
in female primate mating systems, Ethiopia, 167, 294
155–60, 157, 158, 159, 168–71 Afar depression of, 256 fallow deer, 333
linear, 156, 156 Awash Basin of, 252, 256, 264, family, taxonomic, 110
in social behavior, 189–90 282 fats, 129
transitive, 156, 156 Konso-Gardula in, 307 fat-tailed dwarf lemurs, 118
dominance matrix, 156, 156 Omo-Kibish Formation of, 336, Fayum depression, Egypt, 233, 233,
dominant alleles, 30 336, 373 235, 237
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A28
A28 Index
fecundity, 11–12 food sharing and, 288–89, 289, gelada baboons, 119, 167, 167, 203
femur, 249, 250, 250, 265 290, 291–92, 421 gene flow, 87–91, 88, 89, 90, 395
fertility, 445 life expectancy and, 292 between Darwin’s finches, 87, 88,
Fessler, Daniel, 426 by modern humans, 287–88, 287, 92–94, 92, 93
finches, 22 353–54 character displacement and, 93–94
adaptive radiation of, 96, 96 sexual dimorphism and, 292 hominid, 338
characters of, 10, 10, 12, 18–19, sexual division of labor and, 288– migration patterns and, 402–4,
18, 21–22 89, 291, 291 404
correlative characters of, 72–74 slow maturation and, 291–92 reinforcement and, 93–94
courtship displays of, 93 by Stone Age peoples, 376–77 gene frequency:
evolution of, 18–19, 18, 87–88, foramen magnum, 252, 262 calculating, 55–60, 57, 58, 394,
91–94, 92, 93 fossas, 138 394
as example of natural selection, 6– fossil record, 21, 86, 118 equilibrium and, 64–65
10, 176 anatomically modern humans in, evolution and, 63–64
gene flow between, 87, 88, 92–94, 343, 380, 380 genetic drift and, 75–79
92, 93 ancestral characters in, 107 natural selection and, 59–60, 63–
genetic drift and, 79 on anthropoid evolution, 231–37, 64
honeycreeper, 96 233, 234, 242–44, 242 random mating and, 55, 59
interspecies matings among, 89 climate change in, 221–23, 223 sexual reproduction and, 55–57
large ground (G. magnirostris), continental drift in, 221–23, 222 stability reached by, 58
18–19, 18, 19, 89, 89, 91–92, 93 dating methods and, 225–27 genes, 25, 72–73
medium ground (G. fortis), 7–10, environment in, 221–23 on chromosomes, 28–29, 33
7, 12, 18–19, 18, 71, 89, 89, 91– footprints in, 259–60, 260 crossing over of, 36
92, 93 H. ergaster in, 307, 312–13, 318 DNA in, 37–40
morphology of, 9, 18–19, 18, 54, illusion of progress in, 221 enzymes regulated by, 40, 41
61–64, 62, 63, 64, 64, 71 incompleteness of, 238–39, 238, FOXP2, 391–92
reproductive isolation of, 93, 93 239 heterozygous, 29
small ground (G. fuliginosa), 89, missing links in, 238–39, 238, 239 homozygous, 29
89, 92, 93, 93 Neanderthals in, 325 linkage and, 33
species boundaries and, 89, 91 Old World monkeys in, 243–44 mitochondrial, 359–60, 365, 366–
fish, 150, 152, 377 primates in, 228–30, 233, 236 67
eye of, 3, 17 rates of evolutionary change in, mutation of, 65
Fisher, R. A., 61, 61, 64 17–21 natural selection and, 60
fixation, of populations, 78 tools in, 282, 285 pleiotropic effect of, 72, 200–201
flakes (tool), 283, 283, 294, 297, foxes, 355 in populations, 54–55
312, 324–25 FOXP2 gene, 391–92 positively selected, 346–47
cortex of, 284–85, 284 Fragazy, Dorothy, 434, 435 in recombination, 32–36
Flatz, Gebhard, 397 France, 343 regulatory, 347
Fleagle, John, 243 Combe Grenal in, 334 sizes of, 50
Flores, Indonesia, 326–28, 327 Grotte du Renne in, 381–82 structural, 347
Florisbad cranium, 336, 336 La Ferrassie, 344 genetic distance:
Foley, Rob, 337 Solutrean tool tradition in, 351–1, between Neanderthals and
folivores, 130, 133, 134, 149, 234 351 anatomically modern humans,
digestive system of, 133, 134–35 fraternal polyandry, 456–58, 456, 371–72, 371
food, distribution of, 127–35, 127 457 migration from Africa and, 403–4,
food sharing, 421 fraternal (dizygotic) twins, 405–6 403, 404
among chimpanzees, 279–80, 279, Freud, Sigmund, 426 phylogeny and, 107–9, 108, 112
280, 290, 291 frugivores, 130, 134, 134, 234 genetic diversity, 386–415, 387, 389,
hominids and, 279–80, 301, 303 digestive system of, 133 391, 393, 394, 403, 404, 405, 406,
hunting and, 290, 291 fruit flies (Drosophila), 28, 49, 345 407, 408, 409, 411
meat eating and, 289, 290 among groups, 389, 395–404,
by modern humans, 288–89, 289, galagos, 124, 129, 131, 230 398, 400, 401, 404, 406–9, 407,
290, 291, 291 Galápagos Islands, 6–8, 10–11, 18, 408, 409, 413–14, 413
sexual dimorphism and, 292 21, 64, 73, 91–92, 96 in complex phenotypic traits, 404–
foraging, 421 gametes, 25, 29, 31, 56 9, 405, 406, 407, 408, 409
by apes, 287, 287 haploid, 31 explanations of, 386–89, 387, 389
brain size and, 291–92 meiosis and, 28 influenced by single genes, 390–92
by chimpanzees, 287, 287 Garber, Paul, 167 migration patterns reflected in,
EEA and, 421 Gebo, Daniel, 240–41 402–4, 404
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A29
Index A29
modernization and, 408–9, 408 gibbons, 98, 99, 99, 110, 110, 111, griffon vultures, 299
race and, 409–15, 411, 412 125, 148, 243 grooming, 436, 436
within groups, 389, 392–95, 393, ecology of, 137 altruism and, 176–77, 186–88,
394, 405–6, 405, 406, 413–14, mating system of, 165–66, 166 186, 193–95
413 social behavior of, 185 Grotte du Renne, France, 381–82
see also variation in taxonomy of living primates, group selection, 180
genetic drift, 55, 75–79, 75, 76, 77, 125 guanine, 38, 39
78, 94, 100, 389, 401–2, 401 white-handed, 166 guenons, 233
in adaptation, 79 Gibson, Katherine, 203 guinea pigs, 233
and anatomically modern humans, Gilgil, Kenya, 189 gum and gummivores, 129, 130
361–62, 362 Gingerich, Philip, 19 gymnosperms, 227
finches and, 79 glacial periods, 306, 319, 321, 329–
gene frequency and, 75–79 30 Hadza, 75, 288, 288, 289, 299–300,
isolated populations and, 76–78, glucose, 41, 46–47, 48, 49 461, 461
78 glutamic acid, 42 Haile-Selassi, Yohannes, 252
maladaptation and, 79 golden bamboo lemurs, 145 Haldane, J. B. S., 61, 61, 64
rate of, 76–79 golden lion tamarins, 145, 191, 191 Ham, 199
genetic markers, 345, 391–92 golden monkeys, 187 hamadryas baboons, 161
genetics, 24–66 Gombe Stream National Park, Hamilton, W. D., 177, 180, 192
in adaptation, 59–60, 61–63, 73– Tanzania, 94, 158, 159, 160, 277, Hamilton’s rule, 182–84, 183, 188,
74, 75–79, 77, 78, 82 277, 279–80, 290, 336, 336, 422, 189, 192, 456
blending in, 21–22, 64–65 435–36, 436 hand axes, 310–12, 310, 313, 318–
cell division in, 27–28 Gomes de Oliveira, Marino, 435 19, 318, 323, 324
of continuous variation, 60–64 Gondwanaland, 222 Hanihara, Tsunehiko, 363
in evolution, 54–55 Goodall, Jane, 435 Hanuman langurs, 125, 134, 138,
in evolution of behavior, 68–71, gorillas, 86, 86, 96, 98–99, 100, 112, 157, 157, 159, 168, 169, 172
69, 70 118, 126, 131, 131, 154, 155, 241, haploid cells, 28, 29
Mendelian, 24–28 243, 246, 265 haploid gametes, 31
modern synthesis of, 60–61 classification of, 110–12 haplorhines, 126
molecular, 37–50 ecology of, 136, 186 haplotypes, 399–401, 400, 403
in origin and spread of foraging by, 287 Hardy, G. H., 58
anatomically modern humans, language-trained, 429–30, 430 Hardy-Weinberg equilibrium, 58, 59,
360–73, 361, 362, 366, 368, 371 mating system of, 155, 163, 164, 62, 62, 391, 391, 392
population, 53–60, 57, 58, 64 172 Hare, Bryan, 211–12
recombination in, 32–36 morphology of, 86, 86 Harpending, Henry, 448
gene trees, 366–67, 366, 367, 368, mountain, 126, 145, 172 HAR (highly accelerated region)
370 phylogeny of, 98–99, 98, 99, 100, sequences, 348
genome, 33 100, 101, 108, 108, 110, 110, Harris, Christine, 453
evolutionary constraints and, 80, 111 Harvey, Paul, 164
82 reproductive isolation of, 87–88 Haselton, Mark, 450–51
human, 345–46 sexual dimorphism in, 261 Hauser, Marc, 195
in mice, 345–46 grammar, 427–29 hemoglobin, 42, 43, 390, 395, 396
genotypes, 30–31, 30, 33, 35, 53, Grand Canyon, complexity of, 4, 4 Henshilwood, Christopher, 379
65–66, 363, 364, 370 grandparental care, 462–63, 462 Henslow, William, 5
alternative, 54 Grant, Peter, 7–9, 11, 18–19, 21, 62, Herero, 448
body weight and, 388 72, 74, 89 heritability, 21–22, 404
frequency of, 54–58 Grant, Rosemary, 7–9, 11, 18–19, 21, Herrmann, Esther, 213
random mating and, 59 62, 72, 74 Herto, Ethiopia, 373, 374
sexual reproduction and, 57–59 Gravettian tool tradition, 352 heterozygous alleles, 30–31
genotypic frequencies, 54–55 gray langurs, 177 hidden variation, 66–68
genus, 110, 110 great apes (Pongidae), 98–99, 99, Hill, Kim, 286, 288
Geospiza: 110, 110 hippopotamuses, 294, 329, 333
G. fortis (medium ground finch), evolution of locomotion in, 100, Hiwi, 288, 288, 289
7–10, 7, 12, 18–19, 18, 71, 89, 100, 101 Hoberg, Eric, 313
89, 91–92, 93 life history of, 202 home ranges, 135
G. fuliginosa (small ground finch), see also specific species hominids (Hominidae), 221, 246–
89, 89, 92, 93, 93 Great Britain, 19 303, 249, 250, 254, 257, 258, 262,
G. magnirostris (large ground Greenfield, Patricia, 430–31 264, 266, 268, 269, 270, 271, 272,
finch), 18, 18, 89, 89, 91–92, 93 Greenland, 329, 329 274, 275, 276, 278
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A30
A30 Index
in Africa, 240–44, 251, 252, 252, morphology of, 317–18 humans, anatomically modern, 342–
253–70, 254 phylogeny of, 338–39, 339 84
arboreality of, 261 skull of, 317, 317, 330–31 adaptations of, 220
bipedalism and, 247, 249, 252–53, toolmaking by, 318–19 in Africa, 342–43, 358–59, 363–
255, 259, 261–62, 266, 273–75, Homo ergaster, 285–86, 305, 306– 64, 367, 368, 370, 373–74
274, 275 17, 331 art and symbolic expression of,
brain of, 247, 262, 264, 269, 273 basicranium of, 332 349, 357–58, 357, 358, 379–80,
climate change and, 246–47, 275 brain of, 305, 309, 323 379
convergence in evolution of, 271 classification of, 314 in Australia, 342, 349–50
domestic lives of, 301, 302 climate and, 308–9 behavior of, 343, 376–83
in Eurasia, 337–38 development of, 310 bipedalism and, 220
fire use by, 314–15 and emergence of modern humans, body hair loss in, 274
food sharing and, 279–80, 301, 305–6, 309, 321–22, 339 brain of, 21, 205
303 fire use by, 305 and chimpanzee lineages, 345–47
foraging by, 263, 292–301 in fossil record, 307, 312–13, 313 clothing of, 354–55
gene flow and, 338 hunting by, 305, 313, 323–24 culture and, 247
home bases of, 301, 302 as meat eater, 312–13 ecological range of, 349
hunting by, 276, 277–78, 277 morphology of, 309–10, 309, 313, environmental adaptations of,
at kill sites, 297 317–18, 317 354, 355
of late Pleistocene, 324–31, 325 phylogeny of, 338–39, 339 evolution of locomotion in, 100,
of Lower Pleistocene, 306–17 postcranial skeleton of, 309–10, 101
maturation rate of, 263–64 312 in fossil record, 343, 380, 380
of middle Pleistocene, 319–24, running ability of, 310 genetic drift and, 346, 361–62,
338–39 sexual dimorphism of, 310 362
modern humans compared to, skull of, 307–8, 307, 308, 317, genetics and spread of, 358–75,
247, 250, 259 317 361, 362, 366, 368, 371
Oldowan, 283, 285–86, 292–93, teeth of, 308, 313 genetic variation in, 359, 360–64,
295 toolmaking of, 305, 318, 348 360, 361, 362, 397–98, 398
parallelism in evolution of, 271, world distribution of, 307, 315, genomes of, 345
272 315, 318–19 hunting by, 350–52
phylogeny of, 270–71, 271, 272 Homo floresiensis, 326–28, 327 injuries and disease of, 356–57
sagittal crest of, 264, 264 Homo habilis, 254, 285, 286, 307 knee joint of, 250
as scavengers, 298–300 Homo heidelbergensis, 315, 322–24, language and, 220
sexual dimorphism in, 261, 280, 322, 336, 338, 348 life expectancy of, 355–56
303 in Africa, 322–24 in Middle East, 374
skulls of, 247, 252, 256–57, 258, in Eurasia, 322–24, 322, 325, in migration from Africa, 343,
261–62, 262, 264, 264, 268, 268 328 367, 368, 369–70, 369, 403–4,
social organization of, 279–80 H. erectus in coexistence with, 403, 404
teeth of, 247, 251, 252, 253, 255, 325–26, 325 morphology of, 343–45, 358–59
255, 257, 258, 262, 262, 263, phylogeny of, 338–39, 339 mutations in, 360–64, 362, 365
267, 268 skull of, 322–23, 322, 330 Neanderthals and, 370, 371–73,
tool use by, 278, 278, 279, 285–86 homologous characters, 105 371
traits distinguishing modern homologous chromosome pairs, 28 origin and spread of, 358–75
humans from, 247 homology, reasoning by, 117 phylogeny of, 363–64, 366–67,
hominoids (Hominoidea), 110, 110, homozygous alleles, 29, 30, 31, 32, 366, 368
111, 221 33 plant foods and, 353–54
emergence of, 240–44, 241 honeycreeper finch, 96 postcranial skeleton of, 344
phylogeny of, 97–102, 98, 99, horses, 219, 324, 330, 333, 353 religion and, 349
100, 101, 102 Howieson’s Poort tool industry, 378, ritual burial by, 350, 356, 357
Homo, 221, 247, 253, 268, 271, 272, 378 sexual dimorphism in, 261
421, 474 howler monkeys, 119, 145, 280 shelters and communities of, 354–
H. sapiens, see humans, mating system of, 169, 168, 190– 55, 354
anatomically modern; humans, 91 skull of, 266, 343, 344, 344
modern; Neanderthals red, 168, 190–91 of Stone Age, 376–82
Homo erectus, 307, 317–19, 326–27, Hoyle, Fred, 13 toolmaking of, 220, 349–50, 374–
327, 338–39 Hrdy, Sarah Blaffer, 171, 173 75
in Eurasia, 315, 320–21, 325–27 Hublin, Jean-Jacque, 381 traits distinguishing hominids
fire use by, 318 Hugo, 201 from, 247
H. heidelbergensis in coexistence human chorionic gonadotropin, 464 transition in behavior of, 376–83
with, 325–26, 325 human revolution, 343, 376, 382–83 variation in, 360–64, 361, 362
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A31
Index A31
A32 Index
Hamilton’s rule of, 182–84, 183, langurs, 118, 119, 177, 210 mutation and, 66
192 ecology of, 134, 138 in recombination, 33, 36
parent-offspring conflict and, 191– Hanuman, 125, 134, 138, 157, locomotion, evolution of, 100, 100,
92 157, 159, 168, 172 101, 250
phenotype and, 184 mating system of, 157, 158, 168, Lomas Barbudal National Park,
polyandrous marriage and, 457– 172, 173 Costa Rica, 433
58 large ground finch (Geospiza long-tailed macaques, 137, 207
in primates, 184–91, 184, 185, magnirostris), 18–19, 18, 89, 89, Lonsdorf, Elizabeth, 436
186, 187, 188, 189, 190, 191– 91–92, 93 Lordkipanidze, David, 316
92, 193 larynx, 333 lorises, 121, 125–26
in social behavior, 180–92, 183, Later Stone Age (LSA), 376–77 teeth of, 131, 132
184, 185, 186, 187, 188, 189, Laurasia, 222 Lorisiformes, 125–26
190, 191, 193 layer (stratum), 226 LSA (Later Stone Age), 376–77
Kipsigis, 454–56, 454, 455 Leakey, Jonathan, 268 Lucy, 256–57, 256, 257, 259–62,
Klasies River Mouth, South Africa, Leakey, Louis B., 267, 268 269
373, 376–77 Leakey, Mary, 256, 259, 267, 267, Lupemban tool industry, 378, 378
Klein, Richard, 308, 333, 338, 339, 268, 294 luteinizing hormone, 464
375–76, 377, 380, 382, 383 Leakey, Meave, 255, 268
knapping, 283 Leakey, Richard, 269, 307 macaques, 125, 140, 185, 185, 207,
knuckle walking, 100, 100, 101, 275 Lee, Phyllis, 161 209, 209
Koenig, Andreas, 157 Lee-Thorpe, Julia, 263 bonnet, 153, 184, 209–10, 209
Köhler, Meike, 242–43 Lemuriformes, 123, 125–26 brain of, 205
Köhler, Wolfgang, 204 lemurs, 121, 138, 220, 230 ecology of, 137
Komodo dragons, 327 endangered species of, 143 Japanese, 171, 434, 434
Konso-Gardula, Ethiopia, 307 fat-tailed dwarf, 118 long-tailed, 137, 169, 207
Koobi Fora, Kenya, 284, 285, 294, golden bamboo, 145 mating system of, 157, 168
309, 312 mouse, 118, 125–26 social behavior of, 184, 188, 189,
kudus, 253, 253 ring-tailed, 118, 119, 124 193
!Kung, 288, 288, 406, 445 teeth of, 131, 132 toque, 157
leopards, 138, 139–40, 139, 140, McBrearty, Sally, 376, 377, 380
La Chapelle-aux-Saints, France, 334, 299, 300, 429 McGrew, William, 278
335 lesser apes (Hylobatidae), 99 McHenry, Henry, 271, 273
lactation, 80, 103, 104, 116, 128, Levallois tool technique, 324, 324 Machigenga, 260
161, 288 Levine, Nancy, 456 McLarnon, Ann, 310
infanticide and, 171–72 Levi-Strauss, Claude, 426 MacLatchy, Laura, 240–41
parental care and, 152–53 lexigrams, 430, 430 macroevolution, 86
lactose, 46–47, 48, 49, 395–98, 397, Liang Bua, 326 Madagascar, 138, 143, 145
398, 401, 410, 412 Lieberman, Daniel, 310 Magdalenian tool tradition, 352, 353
Laetoli, Tanzania, 256–57, 259–60, Lieberman, Debra, 426 magnetic fields, 227
336 life expectancy: Mahale Mountains, Tanzania, 436,
Lahr, Marta, 337 of anatomically modern humans, 436
Laitman, Jeffrey, 332–33 355–56 Makapansgat, South Africa, 262
Lake Mungo, Australia, 349 foraging and, 292 maladaptation, 79
Lake Turkana, Kenya, 265, 265, 268, of Neanderthals, 334–35, 335, malaria, 393, 393, 395, 396, 399,
268, 269, 269, 307, 309, 309, 312 355–56 412
Laland, Kevin, 203, 204, 205 life history theory, 199–202, 202 mammals, 116, 150, 287
Lancaster, Jane, 286, 459 limpet (Patella sp.), 15 ability to digest lactose in, 395
language, 220, 247, 382 Lincoln Park Zoo, Chicago, 154 adaptive radiation of, 96
as adaptation, 382, 430 linkage, genes and, 33 brain of, 205
brain and, 382 linked loci, 33 placental, 16
capacity for, 430 Linnaeus, Carl, 97, 117 sexual reproduction in, 149
culture and, 426–27 lions, 138, 139, 162, 171, 199, 219, therapsid ancestors of, 218–19,
genetic disease and, 390–92, 391 298, 298, 299, 302 220
of H. ergaster, 310 lizards, 111, 327 mammoths, 323–24, 353, 354
innate knowledge of, 428–29 local adaptations, 79–80 woolly, 330
learning of, 428–29 loci, 33 mandible, 131
of Neanderthals, 332–33 continuous variation and, 60–64 mangabeys, 234, 234
sign, 429–30, 430 distribution of alleles and, 54 Margulis, Susan, 154
Language Instinct, The (Pinker), 333 linked and unlinked, 33, 36 Marlowe, Frank, 461
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A33
Index A33
marmosets, 124, 124, 129, 131, 191, meiosis, 27–28, 29, 30, 35, 37 intelligence of, 206–7, 209, 210,
233, 235 Mellars, Paul, 375 212–13, 213
mating system of, 158, 164, 166– memes, 440–41 mating systems of, 157, 164
67 menarche, 455 New World, see New World
marsupials, 16, 116, 219 Mendel, Gregor, 24–25, 25, 37, 55 monkeys
Martin, Robert, 102, 119, 238–39 Darwinian genetics and, 60–61 Old World, see Old World
Masai, 406 experiments of, 24–28, 26, 29–30, monkeys
mate defense territoriality, 138 30, 31, 32–33 reproduction of, 11–12, 11
mate guarding, 419 principles of, 25, 28, 390–91, sociality in, 141, 205
evolution of, 68–71, 69, 70 391 at typewriters, 13–14
by male primates, 165–66, 166 Menzel, Emil, 210–11, 211 see also specific species
maternal investment, 161 Mesozoic era, 218–19, 219 monogamous social groups, 142, 142
mating effect, 173 messenger RNA, 44, 45 monozygotic (identical) twins, 405–6,
mating systems and mate selection, Mexico, 19, 122 406
148–75, 169, 170, 443–44 mice, 345–46 Morgan, T. H., 25, 29
and age, 448, 449 microevolution, 85, 91 Mormons, 401–2
children and, 459–70 microliths, 376 Morocco, 379
of chimpanzees, 159, 160 Middle Stone Age (MSA), 376–82, Moroto, Uganda, 240–41
commitment in, 450–51, 451 378, 379 Morotopithecus bishopi, 240–41,
culture and, 451–52, 460 Milton, Katherine, 203 240
dominance hierarchies in, 155–60, mind, theory of, 210 morphology, 50
157, 158, 159, 168–71 mineralization, 223 of anatomically modern humans,
female mate choice in, 173, 173 minerals, 129–30 343–45
female reproductive strategies in, minnows, 19–20 behavior and, 68
152–60, 157, 158, 159 minor marriages, 425, 425 chimpanzees, 86, 86, 346
inbred, 422–26, 423, 425 Miocene epoch, 223, 240–44, 246 of finches, 9, 18–19, 18, 54, 61–
infanticide and, 171–73 climate change in, 246–47 63, 62, 63, 64, 64, 71
intersexual selection in, 163, 164 great ape adaptation in, 242–44, of gorillas, 86, 86
male strategies in, 161–65, 162, 242 of H. erectus, 317–18
163, 164, 166, 190–91 miscarriages, 464, 464 of H. ergaster, 309–10, 309, 313,
male tactics in, 165–66, 167–71, Miss Waldron’s red colobus monkey, 317–18, 317
167, 168 143 natural selection and, 89, 95
and number of partners, 448–49, Mitani, John, 194, 277 of Neanderthals, 328–29, 331–32,
450 mitochondria, 359, 360, 365, 366–67 331, 332
parental care in, 150–52, 150, mitochondrial DNA (mtDNA), 360– physical constraints on, 80–82
151, 153, 459–62, 460, 461 73, 366, 368, 369, 371 sexual selection and, 165, 165
paternal care in, 173 mitochondrial Eve, 365, 366–67, 366 species and, 86
of primates, 148–75, 151, 153, mitosis, 27–28, 29, 37 see also sexual dimorphism
157, 158, 159, 161, 163, 164, mode 1 tools, 283, 285, 311, 318–19, mountain gorillas, 126, 143, 145,
165, 166, 167, 168 337, 337 172
psychology in, 443–53, 446, 447 mode 2 tools, 311, 318–19, 318, 337, mouse lemur, 118, 125–26
sexual dimorphism in, 163–64, 337 Mousterian tool industry, 333, 352,
163, 164, 165, 172 mode 3 tools, 324, 324, 333, 337, 375, 377
sexual selection in, 161–65, 162, 337, 376 Moyà-Solà, Salvador, 242–43
163, 164, 165, 166 mode 4 tools, 351, 352 MSA (Middle Stone Age), 376–82,
social behavior in, 143 mode 5 tools, 376 378, 379
social consequences of, 453–58, molars, 131, 131, 133, 233, 234, Mumbwa Cave, Zambia, 379
454, 455, 456, 457 234, 242, 251, 252, 262, 268, 308, Murdock, George Peter, 456
matrilineage rank, 189 331, 331 muriquis (woolly spider monkey),
Matsuzawa, Tesuro, 437 Moldova, 354 145
maxilla, 131 molecular clocks, 108–9 musk oxen, 330
May, Robert, 423 molecular genetics, 37–50 mutation, 55, 109, 192, 382, 388,
Mayr, Ernst, 11, 61 moles, 97, 98 392, 399, 401
meat-eating, 294–95, 294 Mongolia, 350 in anatomically modern humans,
food sharing and, 288–89, 290 monkeys, 119, 121, 125, 130, 135, 360–64, 362, 365
by H. ergaster, 312–13 135, 139–41, 202, 273 loci and, 66
medium ground finch (Geospiza early evolution of, 233, 233, 235, parental care and, 151
fortis), 7–10, 7, 12, 18–19, 18, 71, 235 rate of, 65–66
89, 89, 91–92, 93 infanticide by, 125, 125 variation and, 65–66, 66, 388–89
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A34
A34 Index
Index A35
Pääbo, Svante, 370, 371, 372 optimal, 79 fraternal, 456–58, 456, 457
Packer, Craig, 161 sexual reproduction and, 58 of Nyinba, 456–58, 456, 457
pair-bonding, 165–66 phenylketonuria (PKU), 54, 55–57, polygyny, 142–43, 142
Paleocene epoch, 228 59–60, 73 of Kipsigis, 454–56, 454, 455
paleontology, methods, 223–27 phonemes, 427 polymorphism, 393, 394, 394, 399–
Paleozoic era, 219 photosynthesis, 263 401
Paley, William, 4 phylogenies, 86–112 Pongidae, see great apes
Palombit, Ryne, 166, 173 of anatomically modern humans, Pope, Teresa, 168, 190
Pan, 110, 110 363–64, 366–67, 366, 368 population genetics, 53–60, 57, 58,
Pangaea, 16, 222–23, 227 ancestral characters in, 104–7, 64
Pan paniscus, see bonobos 105–6 porcupines, 233
Pan troglodytes, see chimpanzees, of chimpanzees, 98, 99, 100, 101, porphyria variegata, 402
common 108, 108, 110–11, 110, 111 positively correlated characters, 72
Paraguay, 287, 287 classification and, 109–12, 111 postcranial skeleton:
parallel evolution, 271, 272 in comparative method, 100, 102 of A. africanus, 262
Paranthropus: convergence in, 104–5 of anatomically modern humans,
P. aethiopicus, 253, 265–66, 266, genetic distance in, 107–8, 108, 344
271, 272, 285, 309 112 of chimpanzees, 249
P. boisei, 253, 267, 271, 272 gene trees and, 366–67, 366, 368 of Eocene primates, 231
P. robustus, 253, 266–67, 267, of H. erectus, 338–39, 339 of H. ergaster, 309–10
271, 276 of H. ergaster, 338–39, 339 of P. robustus, 260
parapatric speciation, 95, 95 of H. heidelbergensis, 338–39, 339 potassium-argon dating, 225–26
parapithecids, 233, 235 of hominids, 270–71, 271, 272 pottos, 131, 140
parenting, 459–70 of hominoids, 97–102, 98, 99, Potts, Richard, 302–3, 312
adoption and, 466–70, 468, 100, 101, 102 Povinelli, Daniel, 211
469 of K. platyops, 271 Precambrian era, 219
culture and, 458–59, 460, 461 of modern humans, 98, 100, 104– predation and predators, 287
infanticide and, see infanticide 5, 104, 107–12, 108, 110, 111 alarm calling and, 177, 178–79,
in mating systems, 150–52, 150, of Neanderthals, 338–39, 339 179, 180–81, 181
151, 172–73, 459–62, 460, 461 of primates, 100–102, 100, 101 by chimpanzees, 140, 277, 277
mutations and, 151 reconstruction of, 99–109, 105–6, kill sites and, 296–97
natural selection and, 151 108 by leopards, 139, 139, 140
primates and, 150–52, 150, 151, speciation and, 86–112 in primate ecology, 127, 138–41,
153 taxonomy of, 109–12, 110, 111 139
parent-offspring conflict, 191–92 physics, 80–82 scavenging by, 298–300
Parker, Sue, 203 Pickford, Martin, 251 sociality and, 141
Parra, Flavia, 412–13 picks, 310 Predmostí, Czech Republic, 354
pastoralism, 397–98, 397, 398 pigeons, domestication of, 19, 20 premolars, 131, 234, 234, 251, 252,
patas monkeys, 136, 168, 169 pigs, 294 258, 262, 262
Patterson, Francine, 430 Pinker, Stephen, 333, 428, 431 prenatal investment, 463–65, 464
Pax6, 49 PKU (phenylketonuria), 387, 387, primary structure, 42–44, 43
peacock, 162, 162 392 primates, 28, 116–214
peas, Mendel’s experiments with, 25– plastic environment, 419 activity patterns of, 135–36, 135,
28, 26, 29–30, 30, 31, 53, 54 Pleger, Susanne, 20–21 136
Pekinese (dog), 67–68, 67 pleiotropic effects, 72, 200–201 adaptations by, 117–18
Peres, Carlos, 144 Pleistocene epoch, 244, 305–39, 440, in Africa, 122, 122, 143
Permian period, 218 445 altruistic behavior in, 176–80,
Perry, Susan, 169, 171, 210 climate of, 306, 319–21 179, 181
Peru, 166 Late, 324–39 arboreal, 100–101
Petralona cranium, 322, 323, 325 Lower, 306–17 in Asia, 122, 122, 143
pet trade, 145 Middle, 306, 319–24, 335–36, behavior of, 117, 118–21, 135–38,
phenotypes, 30–31, 30, 31, 35, 53, 337–39, 371–72 179, 181, 210–14, 211, 212,
161 Upper, 306, 326 213, 275
blending genes in, 64–66, 65 plesiadapiforms, 228–30, 228, 229 biogeography of, 122
and expression of characters, 63– Pliocene epoch, 243, 306 brain of, 120, 198–99, 202–5,
64 Poeciliopsis, 19–20, 20 205, 208
heritability and, 404 Poland, 352 characters in definition of, 118–21,
in kin selection, 184 Pollard, Katherine, 347, 348 119, 120, 121, 220, 230–31
natural selection and, 60 polyandry, 142, 142, 165, 166 conservation of, 143–45
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A36
A36 Index
culture and, 433, 434 prosimians (Prosimii), 121, 237 red howler monkeys, 168, 190–91
deceptive behavior in, 210–11, brain of, 205 redirected aggression, 207–8
211 diet of, 130 red-tailed hawks, 110
dietary adaptations of, 133 life history of, 202 regulatory genes, 46–49, 47, 48,
dietary requirements of, 127–35, in taxonomy of living primates, 347
129, 130, 134 122–26, 123 natural selection and, 47, 48
digestive system of, 133–34, 133 teeth of, 131, 228 regulatory sequences, DNA in, 39
diversity within order of, 117–18 proteins, 103, 291 reindeer, 330, 353
earliest appearance of, 202 in amino acids, 42–44, 43 reinforcement, 93–94
early evolution of, 227–39, 242– in chromosomes, 37 repressor protein, 47–48, 47, 48
44, 242 mRNA in synthesis of, 46, 50 reproductive isolation, 87, 88
ecology of, 127–41, 129, 130, non-coding DNA in synthesis of, reproductive value, 445
134, 136, 137, 138, 202, 202, 46 resource defense territoriality, 137–
204, 205, 275 in primate nutrition, 128 38, 137, 138
endangered species of, 143 tRNA in synthesis of, 44 Reznick, David, 19
eyes of, 120, 121, 228 see also enzymes rheas, 152
in fossil record, 228–30, 233, 236 Prugnolle, Franck, 363 rhesus macaques, 212
homologous characters of, 117 psychology, evolutionary, 420, 421 rhesus monkeys, 185, 188, 212
inbreeding among, 171, 422 in inbreeding avoidance, 422–26, rhinoceros, 294, 333
intelligence of, 205, 208, 212–13, 423, 424, 425 woolly, 323–24, 330, 353
213 learning predisposition and, 420, ribonucleic acid, see RNA
kin selection in, 184–91, 184, 185, 421 ribosomes, 49
186, 187, 188, 189, 190, 191, reasoning ability and, 421 Richerson, Peter, 431
193 reciprocity and, 421 Richmond, Brian, 273
knuckle walking of, 100, 100, Punnett square, 33, 33, 35, 36, 59, 59 Rieseberg, Loren, 89–90
101 Pusey, Anne, 158, 161 Rightmire, G. Phillip, 338, 339
lactation in, 80 pygmy chimpanzees, see bonobos ring-tailed lemurs, 118, 119, 124
life span of, 154–55, 198–99 pythons, 138, 139 RNA (ribonucleic acid), 40, 44, 46,
mate guarding in, 165–66, 166, 49–50, 360
177–78 Qafzeh, Israel, 374 messenger, 44, 45, 46
mating systems of, 148–75, 151, quadrupedal locomotion, 100 Robbins, Martha, 158
153, 154–55, 155, 157, 159, rock pigeon, 19, 20
160, 161, 162, 163, 164, 165, race concept, 409–15, 411 rock shelters, 334
166, 167, 168, 169, 172, 177– classification schemes for, 411–13, rodents, 236, 326
78, 178 411, 412, 414 Rodman, Peter, 273
nutrition in ecology of, 127–35, culture and, 414–15 Ross, Corrine, 191
129, 130, 134 as flawed, 410–11 Rubin, Eric, 371
parental care in, 150–52, 150, 151 radioactive decay, 225 Rumbaugh, Duane, 430
phylogeny of, 100–102, 100, 101 radiocarbon (carbon-14) dating, 226, Russia, 353, 354, 355
predation and, 127, 138–41, 139 349 Rwanda, 145, 172
ranging of, 135–38 radiometric dating, 225–27
reasons for study of, 116–17 radius, 265 saber-toothed cat, 16, 302
reciprocal altruism in, 193–95, rain shadow, 240 safari ants, 436
194, 195 Ramachandran, Sohini, 363 sagittal crest, 264, 265
sensory abilities of, 120, 121 raptors, 138 sagittal keel, 317, 317
shrew-like ancestor of, 227–28 Rasmussen, Tab, 230 Sahelanthropus tchadensis, 248, 251,
sociality in, 141–43, 142, 158–60, rats, 420 251
160, 203, 208 Reader, Simon, 203, 204, 205 Sahlins, Marshall, 467
taxonomy of, 117, 122–26, 123 reasoning: Sahul, 330
teeth of, 120, 131–32, 131, 132, by analogy, 117 salmon, 199, 353
228, 228, 234, 234, 267 by homology, 117 Saltzman, Wendy, 422
terrestrial, 100–101, 139–40 recessive alleles, 30 sampling variation, 75–76, 77, 78
territoriality in, 136–38, 137, 138 reciprocal altruism, 193, 194, 195, Santos, Laurie, 212
Proconsul africanus, 241 195 Savage-Rumbaugh, Sue, 430
proconsulids, 241 recombination, 32–36 savanna baboons, 129, 171, 438
progesterone, 464–65, 465 red colobus monkeys, 134, 140–41, scapulae, 331
prokaryotes, 44–45 140, 277, 277 scavenging, hunting and, 298–300
propliopithecids, 233, 235 red deer (elk), 19, 162, 333, 333, Schaafhausen, Hermann, 330
Propliopithecus chirobates, 233 353, 353 Schaik, Carel van, 202
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A37
Index A37
Schick, Kathy, 284–85, 293, 311–12 in mating systems, 161–65, 164, competition in, 141–43
Schmitt, David, 449 165 costs and benefits of, 141–43
Schöningen, Germany, 324 morphology and, 161, 165, 165 dominance hierarchies and, 189–
Schulz, Susanne, 139–40 natural selection and, 161–63 90
sea horses, 152, 152, 163 sexual selection infanticide Hamilton’s rule in, 182–84, 183,
sea otters, 435 hypothesis, 171–72, 173 188, 189
secondary compounds, 134 Seyfarth, Robert, 194, 207, 208, 429 kin selection in, 180–92, 183, 184,
Seemanová, Eva, 423 Shanidar Cave, Iraq, 334, 344 185, 186, 187, 188, 189, 190,
selection: Shepard, Alan, 199 191, 193
artificial, 19, 20, 22, 74 shrews, 220, 227–28 in primate group organization,
group, 180 tree, 220 141–43, 142
individual, 11–12 siamangs, 99, 166 in primate mating systems, 143
intersexual, 163, 164 classification of, 110, 111, 125 reciprocal altruism in, 192–94,
intrasexual, 163–64, 163, 164 Siberia, 350 193, 194, 195
stabilizing, 10 sickle-cell anemia, 42, 390, 393, 393, social facilitation, 433, 434
see also natural selection; sexual 399, 410 social intelligence hypothesis, 203,
selection sifikas, 124 204, 206–10, 207, 209
selection-mutation balance, 392 sign language, 429–30, 430 socioecology, 143
Semaw, Sileshi, 252 Sillen, Andrew, 314 Solutrean tools, 351–52, 351
Senegal, 278 Sima de los Huesos, Spain, 328–29, Sommer, Volker, 157, 173
senescence, 200–201, 201 328 “Sonic Hedgehog,” 49
Senut, Brigitte, 251 Simpson, George Gaylord, 61 South Africa, 253, 256–57, 261, 282,
Serengeti National Park, Tanzania, Sivapithecus, 242, 243 336, 379
162 Skelton, Randall, 271 Afrikaners of, 402
sex ratios, 68–69, 69, 70–71 Skhul Cave, Israel, 374 Drakensberg Mountains, 94
sexual dimorphism, 118, 292 skulls: Howieson’s Poort, 378, 378
in A. afarensis, 261 of A. afarensis, 256–57, 258 Klasies River Mouth, 373, 376–77
in anatomically modern humans, of A. africanus, 262, 262 Makapansgat, 262
261 of anatomically modern humans, Sterkfontein, 262, 267, 269
in baboons, 163 266, 343, 344, 344 Swartkans, 293, 293, 307, 314
in chimpanzees, 261 of chimpanzees, 258 South America, 16, 222, 231
competition and, 163–65 endocranial volume of, 257 deforestation in, 144
food sharing and, 292 of H. erectus, 317, 317, 330–31 endangered primates in, 143
foraging and, 292 of H. ergaster, 307–8, 307, 308, primate evolution in, 237, 237
in gorillas, 261 317, 317 primates in, 122, 122
of H. ergaster, 310 of H. heidelbergensis, 322–23, Spain:
in hominids, 261, 280, 303 322, 330 Sima de los Huesos, 328–29, 328
mating systems and, 163–64, 163, of hominids, 247, 252, 256–57, Trinchera Dolina, 323
164, 165, 172 258, 261–62, 262, 264, 264, spears, 324, 350, 351
see also morphology 268, 271 species, speciation, 86–112
sexual reproduction, 55 of modern humans, 326 in adaptation, 5–6, 10, 12–14
blending in, 64–66, 65 of Neanderthals, 328, 328, 330, allopatric, 91–94, 92, 93, 95, 338–
female strategies in, 152–60, 155, 330, 344 39
157, 159 of Paranthropus, 265, 266 asexual, 90–91
genetic frequency and, 55–57 pneumatized, 252 behavior and, 86
hidden variation and, 66–68, 67 of prosimians, 228 biologic concept of, 86–88, 94, 95
male strategies in, 161–65, 169– sutures of, 335 characters in, 98–99
71, 169, 170 slingjaw wrasse, 17 classification of, 96–99, 98, 103–4,
male tactics in, 165–66, 167–71 sloths, 236 109–12
variation and, 58 small ground finch (Geospiza Darwin on, 11, 21, 86
sexual selection, 161–65, 162, 163, fuliginosa), 89, 89, 92, 93, 93 in Darwin’s explanation of
164, 165, 166 soapberry bug (Jadera haematoloma), variation, 21–22
behavior and, 161 68–71, 69, 70, 419 definitions of, 11, 86, 87–91
characters and, 161 social behavior: ecological concept of, 86–91, 90
definition of, 161 aggression and reconciliation in, endangered, 143
dimorphism and, 163–64, 163, 187–88 evolution of, 19, 20
164, 165 altruism in, 179, 181, 183, 184, of finches, see finches
intersexual, 163, 164 185, 186, 187, 188, 189, 190, gene flow between, 87–91
intrasexual, 163–64, 163, 164 193, 194, 195, 195 in individual selection, 11–12
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A38
A38 Index
morphology and, 86 Taiwanese minor marriages, 425, 425 in primate ecology, 136–38, 137,
niches in, 96, 96 tamarins, 129, 131, 195 138
origin of, 91–96 golden lion, 143, 145, 191, 191 resource defense and, 137–38,
parapatric, 95, 95 mating system of, 158, 166–67, 137, 138
and phylogeny, 86–112 167 tertiary structure, 42, 43
pre-Darwinian view of, 11 Tanzania, 130 testes, 86
sympatric, 95, 95 Gombe Stream National Park, 94, Tethys Sea, 232, 240
specific language impairment (SLI), 158, 159, 160, 277, 277, 279– therapsids, 218, 220
390–92, 391, 392 80, 290, 336, 336, 422, 435–36, thermoluminescence dating, 226, 374
speech, 382 436 third-party relationships, 207, 209–
spherical gradient lens, 3, 17 Laetoli, 256, 259–60, 336 10
spider monkeys, 145 Mahale Mountains, 436, 436 Thrinaxodon, 220
woolly, see muriquis Ngorongoro Crater, 162 thymine, 38, 39, 44, 50
spina bifida, 429 Olduvai Gorge, 267, 267, 268–69, tigers, 138
spiny dye murex, 15 282, 294–95, 294, 295, 296, Tishkoff, Sarah, 397
spliceosomes, 49 297–98, 297, 300, 300, 301–2, titi monkeys, 164, 165
Sponheimer, Matt, 263, 276 301, 302, 307 Tobias, Phillip, 257
“sports,” 22 Serengeti National Park, 162 Tomasello, Michael, 211–12
spotted sandpipers, 152 tapeta, 230 Tooby, John, 420–21, 426
squid, 17 tapeworms, 313, 314 tools and toolmaking, 282–303, 283,
squirrel monkeys, 124, 137, 143, 161 taphonomy, 296 435–37
squirrels, 80 tarsiers, 125, 131, 220, 227, 234, Acheulean, 310–12, 310, 318,
Sri Lanka, 350 234 323, 348–49
stabilizing selection, 10 Tarsiiformes, 125–26 of anatomically modern humans,
Stanford, Craig, 277, 290 Tasmania, 17, 330 220, 349–50, 374–75
stereoscopic vision, 121 Taung child, 261–62, 262 animal bones associated with,
Sterkfontein, South Africa, 262, 267, taurodont teeth, 331, 331 294–96, 294, 295
269 taxonomy, 100, 109–12, 110, 111 Aterian, 378, 378
Stone Age, Middle, 373 defined, 101–2 Aurignacian, 351–52, 352, 375,
Strait, David, 273 Taylor, Andrea, 202 375
“strategy,” use of term, 149–50 Tay-Sachs disease, 387, 387, 392, blade, 351, 352, 377
stratum (layer), 226 393 cached, 303
strepsirhines, 126 teeth, 326 Châtelperronian, 375
structural genes, see coding of A. afarensis, 257, 258 of chimpanzees, 86, 206, 278,
sequences, DNA in of A. africanus, 262, 262, 263 278, 279, 287, 433, 434, 435–
subnasal prognathism, 257, 258 of anthropoids, 241–42 37, 436
Sudan, 377, 378 of A. zeuxis, 235 cut marks on, 293–94, 293, 301
Sultan, 204 bones marked by, 297, 297 earliest identified, 282, 285–86
Sumatra, 202 of chimpanzees, 258 in foraging, 293–94
Sundaland, 330 in defining primates, 120, 121, Gravettian, 352
superfamily, 110 131–32, 131, 132 hafted, 324–25
Sussman, Robert, 229 in dietary adaptations, 133 handedness and, 284–85
Sutton, Walter, 28–29 of early primates, 228, 228, 234, of H. erectus, 318–19
sutures, 335 234 of H. ergaster, 305, 318, 348–49
Suwa, Gen, 252 enamel on, 226 of H. floresiensis, 326
Swartkans, South Africa, 267, 293, of H. ergaster, 308, 313 of H. heidelbergensis, 323, 324–25
293, 307, 314 of hominids, 247, 251, 252, 253, hominid, 278, 278, 279, 285–86,
Sweden, 451–52 255, 255, 257, 258, 262, 262, 292–301
Swisher, Carl, 317 267, 268 Howieson’s Poort, 378, 378
sympatric speciation, 95, 95 of Neanderthals, 331, 331 Levallois, 324, 324
synonymous substitutions, 346 of primates, 120, 132, 234, 267 and long distance movement of
syntax, 428 of proconsulids, 242 resources, 352, 378
Systema Naturae (Linnaeus), 117 of prosimians, 131, 228 of LSA, 376
systematics, 101–2 taurodont roots of, 331, 331 Lupemban, 378, 378
cladistic vs. evolutionary, 110–11 zygomatic arch of, 242 Magdalenian, 352, 353
Szalay, Fred, 229 temporalis muscle, 265, 266 mode 1, 283, 285, 311, 318–19,
termites, 291, 293, 293, 436, 436 337, 337
Taieb, Maurice, 256 terrestrial primates, 100–101, 139–40 mode 2, 311, 318–19, 337, 337
Taï Forest, Ivory Coast, 139–41, 140, territoriality, 136–38, 137 mode 3, 324, 324, 333, 337, 337,
169, 170, 277, 278, 279, 279, 436 mate defense and, 138 376
19_Credits & Index [4p]:How Humans Evolved [5e] 11/10/08 12:24 PM Page A39
Index A39